docs: consolidate root documentation and update architecture#279
Conversation
- Move CONTRIBUTING.md and CHANGELOG.md to root\n- Update README.md with localized documentation links\n- Translate Cerebro migration guide to Spanish\n- Refresh architecture guides (EN/ES) for Rust runtime and Cerebro\n- Add developer-focused README to Cerebro module\n- Fix clippy warning and formatting in agent-runtime
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughDocumentation updates (English/Spanish) and MCP schema text/constraints; added local docs build commands; implemented a Storage::timeline API across Surreal/InMemory/Disk storages and hooked it into the mem_timeline tool in Cerebro, plus metadata-merge behavior in mem_update. Changes
Sequence Diagram(s)sequenceDiagram
participant Client
participant CerebroTool as Cerebro (tools)
participant Storage
participant DB as SurrealDB/Backend
Client->>CerebroTool: mem_timeline(memory_id, before, after, include_deleted)
CerebroTool->>Storage: timeline(memory_id, before, after, include_deleted)
Storage->>DB: query all memory records ordered by timestamp
DB-->>Storage: rows -> Vec<MemoryRecord>
Storage-->>CerebroTool: Vec<MemoryRecord> (windowed slice)
CerebroTool-->>Client: { "items": [...records...] }
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Possibly related PRs
Suggested labels
Suggested reviewers
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
📝 Coding Plan
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
✅ Contributor ReportUser: @yacosta738
Contributor Report evaluates based on public GitHub activity. Analysis period: 2025-03-19 to 2026-03-19 |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 16
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.json`:
- Around line 10-33: The posted mem_get_observation JSON Schema is too strict:
update the schema for the "memory_id" property to require a non-empty string
(e.g., add minLength: 1) to match runtime guarantees, and change the
"observation" property to accept any JSON value (remove the restrictive "type":
"object" and "additionalProperties": true and replace with a schema that allows
any JSON value) so it aligns with the runtime behavior that returns a raw JSON
value from the Cerebro tool code that produces observations.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.json`:
- Around line 1-42: The JSON schema for mem_save_prompt advertises a working
tool but the tool currently returns CerebroError::NotImplemented; either
implement the mem_save_prompt handler in the tools module (replace the
NotImplemented return with the actual save flow that returns prompt_id and
status "saved" to match the schema) or mark the schema as
non-production/unimplemented (e.g., update the title/description to state
"draft/unimplemented" and remove/loosen the output contract) so users won’t
expect a working tool; reference mem_save_prompt and
CerebroError::NotImplemented when making the change so the schema and
implementation stay consistent.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save.json`:
- Around line 10-25: The mem_save schema currently allows empty strings for
"scope", "topic_key", and "observation.content"; update the mem_save JSON schema
to enforce non-empty runtime constraints by adding a minLength: 1 (or equivalent
non-empty string constraint) to the "scope" and "topic_key" properties and to
the nested "observation" → "content" property so the schema matches the
validation enforced in modules/cerebro/src/tools.rs.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_search.json`:
- Around line 3-35: The JSON schema for "mem_search" has English user-facing
description fields (the root "description" and property descriptions for
"query", "limit", "scope", "topic_key", "include_deleted"); update those strings
to Spanish translations or append a clear "pendiente de traducción" note so the
/es/ docs are not left in English; ensure you modify the root "description" and
each property's "description" value inside the "input" object (referenced by
property names query, limit, scope, topic_key, include_deleted) to maintain
EN/ES parity.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_end.json`:
- Around line 3-25: The JSON schema in this Spanish docs path uses English
user-facing text: translate the top-level "title" ("mem_session_end"),
"description" ("Mark an active session as completed.") and the field
descriptions for "session_id", "ended_at", and "metadata" into Spanish (or, if
translation is not ready, add a clear "pending translation" note in Spanish) so
EN/ES parity is preserved; update the strings for the properties inside the
"input" object (session_id, ended_at, metadata) accordingly in the
mem_session_end schema.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.json`:
- Around line 3-30: The schema file mem_session_start.json in the /es/ docs
currently contains English text for "title", "description" and field
descriptions (e.g., properties.input.session_id, properties.input.agent_id,
properties.input.started_at, properties.input.metadata); update those
user-facing strings to Spanish translations or add an explicit translation
status note indicating they are pending translation and must match the EN
version; ensure parity with the English schema (same keys and constraints) while
only changing the text values or adding a clear "translation_pending"
field/comment so reviewers know it's intentionally untranslated.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.json`:
- Around line 1-79: The mem_session_summary schema documents a tool that
currently returns NotImplemented at runtime; either remove this schema from the
docs or implement the actual tool so the docs and runtime match. If removing,
delete or exclude the mem_session_summary JSON schema and any references to it
(search for "mem_session_summary") so users cannot call a non-existent endpoint;
if implementing, add a backend handler (service/function) that accepts the
documented input shape (session_id, summary, metadata), persists or processes
the summary, and returns the documented output (session_id, status
"summarized"), then update the docs to confirm the endpoint is functional and
remove the NotImplemented behavior. Ensure the JSON schema in
mem_session_summary.json remains accurate to the implemented contract and update
any tests or examples that reference it.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.json`:
- Around line 1-46: The schema for mem_suggest_topic_key mismatches the
implementation: the schema defines output.candidates (array) and allows only
"scope" in allowed_arg_fields, but the runtime returns candidates_count
(integer) and requires/validates "seed"; update either side to match. Fix option
A: change the implementation to return the full "candidates" array (max 10
strings) and ensure it accepts/validates "seed" and optional "scope" as inputs;
ensure the output contains "topic_key" and "candidates". Or option B: update
this JSON schema (mem_suggest_topic_key) to replace output.candidates with an
integer "candidates_count", and update allowed_arg_fields to include "seed" (and
mark "scope" optional) so the schema reflects the actual implementation. Target
symbols: mem_suggest_topic_key, input.seed, input.scope, output.candidates,
output.candidates_count, allowed_arg_fields.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_timeline.json`:
- Around line 1-72: The schema file for "mem_timeline" contains English
descriptions but lives in the Spanish docs; either move this JSON to the English
docs or localize the descriptive text to Spanish — update the top-level
"description" and all property description strings under "input" (e.g.,
"memory_id", "before", "after", "include_deleted") and any human-facing text in
"output" (e.g., item "summary", top-level "center_id" description if added) to
Spanish so the content matches the /es/ directory, keeping the schema structure
("title": "mem_timeline", "input", "output", required/additionalProperties)
unchanged.
- Around line 27-30: Update the "description" for the include_deleted schema
property to state that returning deleted records requires audit permissions;
specifically edit the include_deleted object's description (the
"include_deleted" property) to mention the security constraint (e.g., "Requires
audit permissions") so API consumers know the backend enforces `"include_deleted
requires audit permissions"`.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json`:
- Around line 10-34: The JSON schema for mem_update is too strict: update it so
"observation" matches runtime (accept arbitrary JSON like serde_json::Value) and
enforce the runtime non-empty requirement for "memory_id". Specifically, change
the "observation" schema in mem_update.json to allow any JSON value (e.g.,
remove the fixed "properties" object and allow any type/additionalProperties or
use a polymorphic/any schema) so it aligns with modules/cerebro/src/tools.rs
which treats observation as serde_json::Value, and add a minLength: 1 constraint
to the "memory_id" string to match the runtime validation that memory_id cannot
be empty. Ensure the schema change keeps additionalProperties true where needed
to permit arbitrary payloads.
In `@clients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.md`:
- Line 39: The doc text incorrectly names the env var as
CORVUS_CEREBRO_AUTH_TOKEN while the code reads CEREBRO_AUTH_TOKEN; update the
Markdown in migration.md (including the occurrences referenced at the other two
spots) to match the actual env var used by the code by replacing
CORVUS_CEREBRO_AUTH_TOKEN with CEREBRO_AUTH_TOKEN so the docs and the
std::env::var("CEREBRO_AUTH_TOKEN") call in the Cerebro config align.
- Line 63: The sentence saying "no se intenta ninguna alternativa con SurrealDB"
is misleading; update the text to state that Cerebro may fall back to a
configured storage fallback and that StorageFallback::RemoteSurreal is an
available option. Reference the configurable `storage_fallback` behavior and
mention `StorageFallback::RemoteSurreal` as a possible fallback so the doc
matches the implementation in the Storage module.
In `@clients/web/apps/docs/src/content/docs/guides/architecture.md`:
- Around line 36-58: The architecture doc currently claims "Via MCP (JSON-RPC)
from the runtime" without clarifying which MCP endpoints are implemented; update
the English and Spanish architecture pages where that phrase appears to either
list the implemented MCP tools (e.g., mem_save, mem_search, mem_stats,
mem_timeline, mem_suggest_topic_key) or add a short note that some endpoints
(mem_save_prompt, mem_session_start, mem_session_end, mem_session_summary,
mem_context) return NotImplemented and are planned/in-progress; ensure the text
change explicitly names those NotImplemented tools so readers know current vs
planned MCP functionality.
In `@modules/cerebro/README.md`:
- Around line 20-23: The README contains two broken relative links to the
Cerebro docs; update the link targets currently pointing to
"../../docs/guides/cerebro/migration.md" and
"../../docs/guides/cerebro/mcp-schema/" to the correct paths
"../../../clients/web/apps/docs/src/content/docs/guides/cerebro/migration.md"
and "../../../clients/web/apps/docs/src/content/docs/guides/cerebro/mcp-schema/"
respectively so the links resolve; locate and replace the two link strings in
modules/cerebro/README.md.
In `@README.md`:
- Around line 180-193: Update the README links that point to non-existent paths
(`docs/index.mdx` and `docs/guides/`) to the actual docs location
`clients/web/apps/docs/src/content/docs/` (or remove the broken links and use a
short reference), and add the missing local build instructions under the "You
can also build and view the full documentation site locally" section: include
the commands `make docs-dev` for running the local dev server and `make
docs-build` for building the documentation site so users know how to run and
build the docs locally.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 8bd32a91-34dd-4621-b03a-af81fe8fd173
📒 Files selected for processing (34)
CHANGELOG.mdCONTRIBUTING.mdREADME.mdclients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rsclients/web/apps/docs/src/content/docs/es/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_context.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_delete.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_search.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_end.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_stats.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_timeline.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.mdclients/web/apps/docs/src/content/docs/guides/architecture.mdmodules/cerebro/README.mdmodules/cerebro/src/lib.rsmodules/cerebro/src/main.rsmodules/cerebro/src/migration/mod.rsmodules/cerebro/src/storage/mod.rsmodules/cerebro/src/storage/surreal.rsmodules/cerebro/src/tools.rsmodules/cerebro/src/tui/mod.rsmodules/cerebro/tests/embedded_storage_test.rsmodules/cerebro/tests/migration_workflow_test.rsmodules/cerebro/tests/storage_config_test.rsmodules/cerebro/tests/tui_non_blocking_tests.rsmodules/cerebro/tests/tui_redaction_tests.rs
📜 Review details
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (3)
- GitHub Check: pr-checks
- GitHub Check: sonar
- GitHub Check: Cloudflare Pages
🧰 Additional context used
📓 Path-based instructions (8)
**/*.{md,mdx}
⚙️ CodeRabbit configuration file
**/*.{md,mdx}: Verify technical accuracy and that docs stay aligned with code changes.
For user-facing docs, check EN/ES parity or explicitly note pending translation gaps.
Files:
CONTRIBUTING.mdmodules/cerebro/README.mdclients/web/apps/docs/src/content/docs/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/architecture.mdREADME.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.md
**/*
⚙️ CodeRabbit configuration file
**/*: Security first, performance second.
Validate input boundaries, auth/authz implications, and secret management.
Look for behavioral regressions, missing tests, and contract breaks across modules.
Files:
CONTRIBUTING.mdmodules/cerebro/tests/tui_non_blocking_tests.rsmodules/cerebro/tests/tui_redaction_tests.rsclients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rsmodules/cerebro/src/tools.rsmodules/cerebro/src/main.rsmodules/cerebro/tests/storage_config_test.rsmodules/cerebro/tests/migration_workflow_test.rsmodules/cerebro/src/migration/mod.rsclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonmodules/cerebro/README.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.jsonclients/web/apps/docs/src/content/docs/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_stats.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_timeline.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_search.jsonmodules/cerebro/src/tui/mod.rsclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_delete.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_end.jsonmodules/cerebro/tests/embedded_storage_test.rsmodules/cerebro/src/storage/surreal.rsclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.jsonREADME.mdmodules/cerebro/src/storage/mod.rsclients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.jsonmodules/cerebro/src/lib.rsclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_context.json
**/*.rs
⚙️ CodeRabbit configuration file
**/*.rs: Focus on Rust idioms, memory safety, and ownership/borrowing correctness.
Flag unnecessary clones, unchecked panics in production paths, and weak error context.
Prioritize unsafe blocks, FFI boundaries, concurrency races, and secret handling.
Files:
modules/cerebro/tests/tui_non_blocking_tests.rsmodules/cerebro/tests/tui_redaction_tests.rsclients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rsmodules/cerebro/src/tools.rsmodules/cerebro/src/main.rsmodules/cerebro/tests/storage_config_test.rsmodules/cerebro/tests/migration_workflow_test.rsmodules/cerebro/src/migration/mod.rsmodules/cerebro/src/tui/mod.rsmodules/cerebro/tests/embedded_storage_test.rsmodules/cerebro/src/storage/surreal.rsmodules/cerebro/src/storage/mod.rsmodules/cerebro/src/lib.rs
clients/agent-runtime/src/tools/**/*.rs
📄 CodeRabbit inference engine (clients/agent-runtime/AGENTS.md)
Implement
Tooltrait insrc/tools/with strict parameter schema, validate and sanitize all inputs, and return structuredToolResultwithout panics in runtime path
Files:
clients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rs
clients/agent-runtime/src/{security,gateway,tools}/**/*.rs
📄 CodeRabbit inference engine (clients/agent-runtime/AGENTS.md)
Treat
src/security/,src/gateway/,src/tools/as high-risk surfaces and never broaden filesystem/network execution scope without explicit policy checks
Files:
clients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rs
clients/agent-runtime/src/**/*.rs
📄 CodeRabbit inference engine (clients/agent-runtime/AGENTS.md)
clients/agent-runtime/src/**/*.rs: Never log secrets, tokens, raw credentials, or sensitive payloads in any logging statements
Avoid unnecessary allocations, clones, and blocking operations to maintain performance and efficiency
Files:
clients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rs
clients/agent-runtime/**/*.rs
📄 CodeRabbit inference engine (clients/agent-runtime/AGENTS.md)
Run
cargo fmt --all -- --check,cargo clippy --all-targets -- -D warnings, andcargo testfor code validation, or document which checks were skipped and why
Files:
clients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rs
clients/agent-runtime/src/{security,gateway,tools,config}/**/*.rs
📄 CodeRabbit inference engine (clients/agent-runtime/AGENTS.md)
Do not silently weaken security policy or access constraints; keep default behavior secure-by-default with deny-by-default where applicable
Files:
clients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rs
🧠 Learnings (13)
📓 Common learnings
Learnt from: CR
Repo: dallay/corvus PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-17T07:28:38.934Z
Learning: Applies to .agents/AGENTS.md : Document agent configurations and capabilities in AGENTS.md
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/main.rs : Preserve CLI contract unless change is intentional and documented; prefer explicit errors over silent fallback for unsupported critical paths
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/{security,gateway,tools,config}/**/*.rs : Do not silently weaken security policy or access constraints; keep default behavior secure-by-default with deny-by-default where applicable
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/{security,gateway,tools}/**/*.rs : Treat `src/security/`, `src/gateway/`, `src/tools/` as high-risk surfaces and never broaden filesystem/network execution scope without explicit policy checks
Applied to files:
CONTRIBUTING.mdclients/agent-runtime/src/tools/shell.rs
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/{security,gateway,tools,config}/**/*.rs : Do not silently weaken security policy or access constraints; keep default behavior secure-by-default with deny-by-default where applicable
Applied to files:
CONTRIBUTING.mdclients/agent-runtime/src/tools/shell.rs
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/**/*.rs : Run `cargo fmt --all -- --check`, `cargo clippy --all-targets -- -D warnings`, and `cargo test` for code validation, or document which checks were skipped and why
Applied to files:
CONTRIBUTING.mdmodules/cerebro/tests/tui_redaction_tests.rsclients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rs
📚 Learning: 2026-02-17T07:28:38.934Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-17T07:28:38.934Z
Learning: Applies to .agents/AGENTS.md : Document agent configurations and capabilities in AGENTS.md
Applied to files:
CONTRIBUTING.md
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/.github/**/*.{yml,yaml} : For workflow/template-only changes, ensure YAML/template syntax validity
Applied to files:
CONTRIBUTING.md
📚 Learning: 2026-02-17T07:28:38.934Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-17T07:28:38.934Z
Learning: Applies to .agents/AGENTS.md : Maintain comprehensive agent metadata including name, description, purpose, and capabilities
Applied to files:
CONTRIBUTING.md
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/**/*.rs : Avoid unnecessary allocations, clones, and blocking operations to maintain performance and efficiency
Applied to files:
CONTRIBUTING.mdclients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rs
📚 Learning: 2026-02-17T07:28:38.934Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-17T07:28:38.934Z
Learning: Applies to .agents/AGENTS.md : Include version information and compatibility details for agents
Applied to files:
CONTRIBUTING.md
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Include threat/risk notes and rollback strategy for security, runtime, and gateway changes; add or update tests for boundary checks and failure modes
Applied to files:
CONTRIBUTING.md
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/main.rs : Preserve CLI contract unless change is intentional and documented; prefer explicit errors over silent fallback for unsupported critical paths
Applied to files:
CONTRIBUTING.mdclients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rsmodules/cerebro/src/tools.rsmodules/cerebro/src/migration/mod.rs
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/tools/**/*.rs : Implement `Tool` trait in `src/tools/` with strict parameter schema, validate and sanitize all inputs, and return structured `ToolResult` without panics in runtime path
Applied to files:
clients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rsmodules/cerebro/src/tools.rs
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/main.rs : Keep startup path lean and avoid heavy initialization in command parsing flow
Applied to files:
clients/agent-runtime/src/tools/mcp/client.rsclients/agent-runtime/src/tools/shell.rsmodules/cerebro/src/main.rs
🪛 LanguageTool
CONTRIBUTING.md
[uncategorized] ~267-~267: The official name of this software platform is spelled with a capital “H”.
Context: ...h trait to extend - Security: See .github/SECURITY.md for r...
(GITHUB)
[uncategorized] ~267-~267: The official name of this software platform is spelled with a capital “H”.
Context: ...- Security: See .github/SECURITY.md for responsible disclosure...
(GITHUB)
clients/web/apps/docs/src/content/docs/es/guides/architecture.md
[grammar] ~36-~36: Cambia la palabra o signo.
Context: ...s siguientes pilares: ### 1. Runtime Reactivo - Tecnología: Rust (Tokio/Async) - **Pro...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_OTHER)
[grammar] ~38-~38: Elimina el sustantivo
Context: ...ares: ### 1. Runtime Reactivo - Tecnología: Rust (Tokio/Async) - Propósito: Eje...
(QB_NEW_ES_OTHER_ERROR_IDS_UNNECESSARY_NOUN)
[grammar] ~38-~38: Elimina el sustantivo
Context: ...** - Tecnología: Rust (Tokio/Async) - Propósito: Ejecución concurrente y de alto rendimi...
(QB_NEW_ES_OTHER_ERROR_IDS_UNNECESSARY_NOUN)
[grammar] ~39-~39: Cambia la palabra o signo.
Context: ...urrente y de alto rendimiento de agentes - Ubicación: Runtime del agente en `clients/agent-ru...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~40-~40: Elimina la preposición
Context: ...tes - Ubicación: Runtime del agente en clients/agent-runtime/ ### 2. Cerebro (Memoria a Largo Plazo) ...
(QB_NEW_ES_OTHER_ERROR_IDS_UNNECESSARY_ADPOSITION)
[grammar] ~42-~42: Corrige la mayúscula.
Context: ...-runtime/` ### 2. Cerebro (Memoria a Largo Plazo) - Tecnología: Rust + Surr...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_ORTHOGRAPHY_UPPERCASE)
[grammar] ~42-~42: Corrige la mayúscula.
Context: ...me/` ### 2. Cerebro (Memoria a Largo Plazo) - Tecnología: Rust + SurrealDB ...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_ORTHOGRAPHY_UPPERCASE)
[grammar] ~42-~42: Aquí puede haber un error.
Context: ...### 2. Cerebro (Memoria a Largo Plazo) - Tecnología: Rust + SurrealDB (Embebido...
(QB_NEW_ES)
[grammar] ~44-~44: Oración con errores
Context: ...** - Tecnología: Rust + SurrealDB (Embebido) - Propósito: Servicio de memoria c...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_MULTITOKEN)
[grammar] ~44-~44: Oración con errores
Context: ...ecnología**: Rust + SurrealDB (Embebido) - Propósito: Servicio de memoria central...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_MULTITOKEN)
[grammar] ~45-~45: Cambia la palabra o signo.
Context: ...iente del agente con soporte para grafos - Integración: Vía MCP (JSON-RPC) desde ...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~50-~50: Cambia la palabra o signo.
Context: ...Tecnología: Rust + Docker (opcional) - Propósito: Ejecución segura de código ...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~51-~51: Cambia la palabra o signo.
Context: ...o no confiable y operaciones del sistema - Comunicación: Gestionada por el runtim...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~54-~54: Cambia la palabra o signo.
Context: ...el runtime reactivo ### 4. Panel del Operador - Tecnología: Astro + Vue 3 - **Propósit...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_OTHER)
[grammar] ~56-~56: Elimina la palabra o signo.
Context: ...ivo ### 4. Panel del Operador - Tecnología: Astro + Vue 3 - Propósito: Observab...
(QB_NEW_ES_OTHER_ERROR_IDS_UNNECESSARY_OTHER)
[grammar] ~56-~56: Cambia la palabra o signo.
Context: ...rador** - Tecnología: Astro + Vue 3 - Propósito: Observabilidad en tiempo real e interve...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~57-~57: Cambia la palabra o signo.
Context: ...dad en tiempo real e intervención manual - Ubicación: clients/web/ ## Lógica de Construcción (Plugins de Conve...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
README.md
[style] ~193-~193: Using many exclamation marks might seem excessive (in this case: 22 exclamation marks for a text that’s 5977 characters long)
Context: ... Contributing Contributions are welcome! Please read our [CONTRIBUTING.md](CONTR...
(EN_EXCESSIVE_EXCLAMATION)
clients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.md
[grammar] ~7-~7: Elimina la palabra o signo.
Context: ...consulta la especificación de Cerebro en https://github.com/dallay/corvus/blob/ma...
(QB_NEW_ES_OTHER_ERROR_IDS_UNNECESSARY_SPACE)
[grammar] ~116-~116: Cambia la palabra o signo.
Context: ...i.enabled(booleano, por defecto false) -tui.event_buffer` (tamaño del búfer de eventos acotado) - ...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~117-~117: Cambia la palabra o signo.
Context: ...r(tamaño del búfer de eventos acotado) -tui.refresh_ms(intervalo de refresco de la UI) -tui....
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~118-~118: Cambia la palabra o signo.
Context: ...esh_ms(intervalo de refresco de la UI) -tui.redact_fields` (lista de denegación para claves sensibl...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~119-~119: Cambia la palabra o signo.
Context: ...sta de denegación para claves sensibles) - tui.max_payload_bytes (límite de carga para datos redactados) ...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_SPACE)
[grammar] ~120-~120: Aquí puede haber un error.
Context: ... (límite de carga para datos redactados) Notas de seguridad: - Los eventos de ll...
(QB_NEW_ES)
[grammar] ~141-~141: Cambia el sustantivo.
Context: ...ort.json \ --target ./cerebro.db ``` Flags opcionales: - --namespace / `--datab...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_NOUN)
[grammar] ~145-~145: Aquí puede haber un error.
Context: ...mbres embebido específico. - --dry-run para calcular recuentos/checksums sin re...
(QB_NEW_ES)
[grammar] ~145-~145: Corrige la minúscula.
Context: ...bres embebido específico. - --dry-run para calcular recuentos/checksums sin realiz...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_ORTHOGRAPHY_LOWERCASE)
🔇 Additional comments (36)
clients/agent-runtime/src/tools/mcp/client.rs (3)
341-341: Formatting-only refactor is safe here.Line 341 and Lines 347-350 keep the same bounded-read behavior (
take(limit + 1)+read_to_end) and preserve truncation detection semantics.Also applies to: 347-350
381-383: Receiver timeout/destructuring change preserves runtime behavior.Lines 381-383 only reflow the
recv_timeoutassignment; error propagation and timeout handling are unchanged.
394-395: UTF-8 conversion refactor is behaviorally equivalent.Lines 394-395 retain the same UTF-8 validation and error context, with no contract change.
modules/cerebro/tests/migration_workflow_test.rs (1)
2-4: LGTM! Import reordering improves clarity.The consolidation of migration imports and early placement of
serde_json::Valueandstd::fsaligns with common Rust import organization patterns.modules/cerebro/src/migration/mod.rs (4)
32-33: LGTM! Idiomatic inline error mapping.The inline chained
map_erris cleaner and more idiomatic than the multi-line closure format.
52-53: LGTM! Consistent error handling style.Matches the formatting improvement in
import_legacy_exportfor consistency.
116-122: LGTM! Formatting improves readability.The function signature and insert statement formatting is clean and follows Rust conventions.
127-128: LGTM! Consistent with the file's formatting style.The inline
map_errchain maintains consistency with the other formatting improvements in this file.clients/agent-runtime/src/tools/shell.rs (1)
450-453: LGTM: Formatting-only changes.The reformatting to single-line expressions doesn't alter test behavior or security properties. The timeout validation logic remains intact.
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_timeline.json (1)
15-25: The schema enforces a maximum of 100 for bothbeforeandafterparameters, which matches theMAX_TIMELINE_ITEMSconstant in the backend implementation.modules/cerebro/src/lib.rs (1)
10-16: LGTM!The re-export formatting changes are consistent and maintain the same public API surface.
modules/cerebro/src/tools.rs (2)
1-9: LGTM!Import reordering maintains the same imports with improved readability.
376-384: LGTM!The simplified error message construction maintains identical error content for not-implemented tools while improving readability.
modules/cerebro/src/storage/surreal.rs (2)
11-11: LGTM!Import reordering maintains the same dependencies.
74-130: LGTM!The formatting changes to UPSERT statements and error mapping maintain identical logic and error messages while improving readability.
modules/cerebro/src/storage/mod.rs (4)
12-12: LGTM!Import reordering maintains the same dependencies.
138-141: LGTM!The multi-line format! call maintains identical error message content with improved readability.
257-260: LGTM!The reformatted predicate maintains identical logic with improved readability.
361-361: LGTM!Function signature formatting maintains the same public contract.
modules/cerebro/tests/embedded_storage_test.rs (2)
2-2: LGTM!Import reordering maintains the same test dependencies.
19-19: LGTM!The single-line function call maintains identical test setup logic with improved readability.
modules/cerebro/tests/storage_config_test.rs (1)
139-142: LGTM!The reformatted error mapping maintains identical error handling logic with improved readability.
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_stats.json (1)
1-34: Schema correctly matches implementation; EN/ES versions exist.The JSON schema accurately reflects the
mem_statsimplementation inmodules/cerebro/src/tools.rs(lines 625-636). Both English and Spanish versions of the schema are present, maintaining documentation parity.CONTRIBUTING.md (2)
267-267: Security policy path is correct. The.github/SECURITY.mdfile exists and the reference in CONTRIBUTING.md is valid.
63-64: Documentation paths are correctly referenced and exist in the repository with EN/ES parity maintained.modules/cerebro/tests/tui_redaction_tests.rs (1)
35-35: LGTM - formatting change only.The assertion was reformatted to a single line with no logic changes. The test behavior remains unchanged.
modules/cerebro/tests/tui_non_blocking_tests.rs (1)
67-72: LGTM - formatting improvements.The
start_tui_taskinvocation andshutdown_tx.sendwere reformatted to multi-line for improved readability. No logic changes.Also applies to: 88-90
modules/cerebro/src/main.rs (1)
2-2: LGTM - formatting and import reordering only.Import statements were reordered and the
start_tui_taskcall was reformatted to single-line. No functional changes to the startup path or initialization logic.Also applies to: 5-5, 37-37
clients/web/apps/docs/src/content/docs/guides/architecture.md (1)
36-58: Architecture changes align with codebase migration.The pillar updates correctly reflect:
- Rust-based reactive runtime (Tokio/Async) in
clients/agent-runtime/- Cerebro long-term memory service in
modules/cerebro/- Sandboxed tools with Docker support
- Operator dashboard using Astro + Vue 3
These changes match the documented migration from Kotlin/Spring to Rust runtime. Based on learnings, ensure the Spanish version (
docs/es/guides/architecture.md) maintains EN/ES parity.modules/cerebro/src/tui/mod.rs (1)
10-10: Formatting-only Rust changes look safe.These edits are non-functional (import/line-wrapping/readability) and do not alter control flow, error handling, or TUI behavior.
Also applies to: 57-57, 204-205, 403-409, 474-481
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_delete.json (1)
1-58: Schema aligns with implementation.The
mem_deleteschema correctly enforces the mutual-exclusivity pattern (oneOf formemory_id/topic_key) and matches the implementation's status enum values and output structure.clients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.md (3)
23-27: Transport security enforcement is accurately documented.The secure-by-default transport policy (https/wss required, http/ws only for loopback) correctly reflects the implementation's scheme validation logic.
55-61: Tool mapping is accurate.The legacy-to-MCP tool aliases (
memory_store→mem_save,memory_recall→mem_search,memory_forget→mem_delete) are clearly documented for migration.
98-126: TUI operational details are well-documented.The optional TUI configuration, security notes (redaction, backpressure), and no-network-ports guarantee provide clear operational guidance.
clients/web/apps/docs/src/content/docs/es/guides/architecture.md (1)
36-58: No action needed. EN/ES documentation parity is maintained. Bothguides/architecture.mdandes/guides/architecture.mdhave received equivalent updates reflecting the Rust/MCP-based architecture design, with matching section structures and content at the same line positions.clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_context.json (1)
1-62: mem_context is not yet implemented—schema verification cannot be completed.The
mem_contexttool explicitly returnsNotImplementedinmodules/cerebro/src/tools.rs(line 384). Since there is no implementation to validate against, the schema alignment with input/output contracts, timestamp format handling, and limit bounds validation cannot be verified. The schema itself is valid JSON, but aligns with a planned feature rather than an existing implementation.> Likely an incorrect or invalid review comment.
- Fix JSON schema constraints and descriptions for Spanish docs\n- Align mem_suggest_topic_key schema with runtime implementation\n- Translate remaining English strings in Spanish documentation\n- Update architecture overview with implemented vs planned tools\n- Fix environment variable name in migration guide\n- Repair broken links in root and module READMEs\n- Add local documentation build instructions
Deploying corvus with
|
| Latest commit: |
0165430
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://77654b99.corvus-42x.pages.dev |
| Branch Preview URL: | https://docs-documentation-improveme.corvus-42x.pages.dev |
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
There was a problem hiding this comment.
Actionable comments posted: 5
♻️ Duplicate comments (3)
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.json (1)
3-4:⚠️ Potential issue | 🟠 MajorDraft label is good, but the schema still requires a success response for an unimplemented tool.
mem_save_promptcurrently returnsCerebroError::NotImplemented, yet this schema still mandatesoutput.prompt_idandoutput.status="saved". That is still a contract mismatch for generated clients.💡 Suggested patch
- "description": "Save a prompt to Cerebro (NotImplemented returning CerebroError::NotImplemented).", + "description": "Borrador/no implementado: actualmente devuelve CerebroError::NotImplemented.", @@ - "required": ["input", "output"], + "required": ["input"],As per coding guidelines,
**/*: "Look for behavioral regressions, missing tests, and contract breaks across modules."Also applies to: 30-41
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.json` around lines 3 - 4, The schema for mem_save_prompt conflicts with the implementation that returns CerebroError::NotImplemented; update the JSON schema for mem_save_prompt so it no longer requires a successful output with output.prompt_id and output.status="saved" — either make those fields optional or add an alternative error response variant that models CerebroError::NotImplemented; ensure references to mem_save_prompt, output.prompt_id, output.status and CerebroError::NotImplemented are handled so generated clients won’t assume a guaranteed "saved" success.clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json (1)
14-17:⚠️ Potential issue | 🟠 Major
observationis still not modeled as raw JSON.The handler accepts this field as arbitrary
serde_json::Value, so arrays, strings, numbers, booleans, andnullare valid inputs. Keeping"type": "object"still rejects payloads the runtime accepts.Proposed fix
"observation": { - "type": "object", - "description": "New observation payload (raw JSON). Replaces the existing one if provided." + "description": "Nuevo payload de observación (JSON arbitrario). Sustituye al existente si se envía." },As per coding guidelines,
**/*: Look for behavioral regressions, missing tests, and contract breaks across modules.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json` around lines 14 - 17, The schema for the "observation" field currently restricts it to "type": "object" but the runtime accepts arbitrary JSON (serde_json::Value); update the mem_update.json "observation" schema (the "observation" property) to allow any JSON value—for example replace the single "type": "object" with a broader schema such as a oneOf/anyOf that includes object, array, string, number, boolean and null (e.g., oneOf with types "object","array","string","number","boolean","null") while keeping the existing description.clients/web/apps/docs/src/content/docs/guides/architecture.md (1)
47-48:⚠️ Potential issue | 🟡 MinorMove
mem_timelineout of the NotImplemented list.The backend dispatch already routes
mem_timelineto a real handler, so listing it under Planned Tools (NotImplemented) is inaccurate. Please move it into the implemented list and mirror the same correction inclients/web/apps/docs/src/content/docs/es/guides/architecture.md.Proposed fix
- - **Implemented Tools**: `mem_save`, `mem_search`, `mem_delete`, `mem_get_observation`, `mem_update`, `mem_suggest_topic_key`, `mem_stats`. - - **Planned Tools (NotImplemented)**: `mem_save_prompt`, `mem_session_start`, `mem_session_end`, `mem_session_summary`, `mem_context`, `mem_timeline`. + - **Implemented Tools**: `mem_save`, `mem_search`, `mem_delete`, `mem_get_observation`, `mem_update`, `mem_suggest_topic_key`, `mem_stats`, `mem_timeline`. + - **Planned Tools (NotImplemented)**: `mem_save_prompt`, `mem_session_start`, `mem_session_end`, `mem_session_summary`, `mem_context`.As per coding guidelines,
**/*.{md,mdx}: Verify technical accuracy and that docs stay aligned with code changes.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@clients/web/apps/docs/src/content/docs/guides/architecture.md` around lines 47 - 48, Move the mem_timeline entry out of the "Planned Tools (NotImplemented)" list and add it to the "Implemented Tools" list (i.e., ensure the Implemented Tools list includes mem_timeline alongside mem_save, mem_search, etc.); apply the same change in the Spanish translation of the same architecture guide so both docs reflect that mem_timeline is implemented and the two lists remain properly formatted and punctuated.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.json`:
- Line 13: The JSON schema still contains English description strings (e.g.,
"Memory ID to retrieve.") that need Spanish localization; locate the description
values containing that exact text and the other two English description fields
in the same file and replace them with their Spanish equivalents (for example
"ID de la memoria a recuperar."), ensuring all "description" values in the
schema are translated consistently into Spanish.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.json`:
- Around line 3-4: Update the schema header for the mem_session_start endpoint
to mark it unimplemented so it matches runtime behavior; specifically change the
schema metadata for the "mem_session_start" endpoint (title "Inicio de sesión de
memoria") to use the same draft/unimplemented treatment used by
mem_session_summary (or remove the mem_session_start schema entirely) so the
header reflects that calls currently return
CerebroError::NotImplemented("mem_session_start").
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.json`:
- Around line 3-4: The Spanish header is fine but the field descriptions for
session_id, summary, and metadata are still in English; update the user-facing
strings for the JSON fields "session_id", "summary", and "metadata" to full
Spanish translations (or replace each with a clear pending-translation note in
Spanish) so the /es/ schema is not mixed-language; also apply the same change to
the other occurrences of these field docs elsewhere in this file (the later
blocks noted in the review) to ensure consistency.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.json`:
- Around line 15-17: The JSON docs incorrectly state that "scope" is used for
collision checks even though the suggestion handler derives suggestions from
"seed" only; update the schema/docs to mark "scope" as reserved/ignored (or
remove the collision-check description) OR implement scope-aware suggestion
logic in the suggestion handler so it actually reads "scope" when
generating/validating suggestions; ensure you update the "scope" field
description in mem_suggest_topic_key.json to reflect the chosen behavior and
keep the "seed" usage consistent with the handler that currently derives
suggestions from seed.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json`:
- Around line 26-29: The schema advertises input.metadata as mergeable but the
mem_update handler never reads input.metadata; update the mem_update
implementation to consume input.metadata from the request and merge it into the
existing memory's metadata (e.g., in the mem_update handler/function where you
locate the current memory by id), performing a shallow/deep merge consistent
with other metadata behavior and persisting the merged metadata, or if you
prefer not to support this yet, remove/mark the metadata property in
mem_update.json as unsupported/reserved and update any related docs/tests;
reference the mem_update handler, input.metadata, and the mem_update.json
metadata entry when making the change.
---
Duplicate comments:
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.json`:
- Around line 3-4: The schema for mem_save_prompt conflicts with the
implementation that returns CerebroError::NotImplemented; update the JSON schema
for mem_save_prompt so it no longer requires a successful output with
output.prompt_id and output.status="saved" — either make those fields optional
or add an alternative error response variant that models
CerebroError::NotImplemented; ensure references to mem_save_prompt,
output.prompt_id, output.status and CerebroError::NotImplemented are handled so
generated clients won’t assume a guaranteed "saved" success.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json`:
- Around line 14-17: The schema for the "observation" field currently restricts
it to "type": "object" but the runtime accepts arbitrary JSON
(serde_json::Value); update the mem_update.json "observation" schema (the
"observation" property) to allow any JSON value—for example replace the single
"type": "object" with a broader schema such as a oneOf/anyOf that includes
object, array, string, number, boolean and null (e.g., oneOf with types
"object","array","string","number","boolean","null") while keeping the existing
description.
In `@clients/web/apps/docs/src/content/docs/guides/architecture.md`:
- Around line 47-48: Move the mem_timeline entry out of the "Planned Tools
(NotImplemented)" list and add it to the "Implemented Tools" list (i.e., ensure
the Implemented Tools list includes mem_timeline alongside mem_save, mem_search,
etc.); apply the same change in the Spanish translation of the same architecture
guide so both docs reflect that mem_timeline is implemented and the two lists
remain properly formatted and punctuated.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: a3054c71-3809-4215-92d6-e6454e8148b3
📒 Files selected for processing (15)
README.mdclients/web/apps/docs/src/content/docs/es/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_search.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_end.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_timeline.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.mdclients/web/apps/docs/src/content/docs/guides/architecture.mdmodules/cerebro/README.md
📜 Review details
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (4)
- GitHub Check: pr-checks
- GitHub Check: sonar
- GitHub Check: pr-checks
- GitHub Check: Cloudflare Pages
🧰 Additional context used
📓 Path-based instructions (2)
**/*.{md,mdx}
⚙️ CodeRabbit configuration file
**/*.{md,mdx}: Verify technical accuracy and that docs stay aligned with code changes.
For user-facing docs, check EN/ES parity or explicitly note pending translation gaps.
Files:
modules/cerebro/README.mdclients/web/apps/docs/src/content/docs/guides/architecture.mdREADME.mdclients/web/apps/docs/src/content/docs/es/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.md
**/*
⚙️ CodeRabbit configuration file
**/*: Security first, performance second.
Validate input boundaries, auth/authz implications, and secret management.
Look for behavioral regressions, missing tests, and contract breaks across modules.
Files:
modules/cerebro/README.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_timeline.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_end.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save.jsonclients/web/apps/docs/src/content/docs/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonREADME.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.jsonclients/web/apps/docs/src/content/docs/es/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_search.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.md
🧠 Learnings (6)
📚 Learning: 2026-02-17T07:28:38.934Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-17T07:28:38.934Z
Learning: Applies to .agents/AGENTS.md : Document agent configurations and capabilities in AGENTS.md
Applied to files:
modules/cerebro/README.md
📚 Learning: 2026-02-17T07:28:38.934Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-17T07:28:38.934Z
Learning: Applies to .agents/AGENTS.md : Include version information and compatibility details for agents
Applied to files:
modules/cerebro/README.md
📚 Learning: 2026-02-17T07:28:38.934Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-17T07:28:38.934Z
Learning: Applies to .agents/AGENTS.md : Maintain comprehensive agent metadata including name, description, purpose, and capabilities
Applied to files:
modules/cerebro/README.md
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/main.rs : Preserve CLI contract unless change is intentional and documented; prefer explicit errors over silent fallback for unsupported critical paths
Applied to files:
clients/web/apps/docs/src/content/docs/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/tools/**/*.rs : Implement `Tool` trait in `src/tools/` with strict parameter schema, validate and sanitize all inputs, and return structured `ToolResult` without panics in runtime path
Applied to files:
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/**/*.rs : Run `cargo fmt --all -- --check`, `cargo clippy --all-targets -- -D warnings`, and `cargo test` for code validation, or document which checks were skipped and why
Applied to files:
README.md
🪛 LanguageTool
clients/web/apps/docs/src/content/docs/es/guides/architecture.md
[grammar] ~45-~45: Agrega una palabra o signo.
Context: ... MCP (JSON-RPC) desde el runtime. - Herramientas Implementadas: mem_save, `mem_searc...
(QB_NEW_ES_OTHER_ERROR_IDS_MISSING_OTHER)
[grammar] ~45-~45: Corrige la mayúscula.
Context: ...C) desde el runtime. - Herramientas Implementadas: mem_save, mem_search, `mem_delet...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_ORTHOGRAPHY_UPPERCASE)
[grammar] ~46-~46: Corrige la mayúscula.
Context: ...c_key, mem_stats. - **Herramientas Planificadas (NotImplemented)**: mem_save_prompt`, ...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_ORTHOGRAPHY_UPPERCASE)
[grammar] ~47-~47: Elimina la puntuación
Context: ...eline`. - Almacenamiento: Soporte multi-modelo que incluye SurrealDB embebido, JSON en...
(QB_NEW_ES_OTHER_ERROR_IDS_UNNECESSARY_PUNCTUATION)
🔇 Additional comments (6)
modules/cerebro/README.md (1)
20-22: Integration links now point to the correct docs tree.Nice fix—this aligns module docs with the current repository documentation layout.
README.md (1)
182-183: Documentation entry points and local docs commands look correct.Good update: EN/ES links are explicit and the local docs workflow is now actionable.
Also applies to: 189-190
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_end.json (1)
4-4: Spanish localization updates are clear and consistent.This improves schema readability without altering validation behavior.
Also applies to: 14-14, 19-19, 23-23
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_timeline.json (1)
29-29: Permission requirement forinclude_deletedis now documented.Great improvement—this makes the auth expectation explicit for API consumers.
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save.json (1)
12-12: Non-empty boundary constraints are correctly enforced.This aligns the schema with stricter runtime expectations for required string inputs.
Also applies to: 17-17, 26-26
clients/web/apps/docs/src/content/docs/es/guides/cerebro/migration.md (1)
39-39: Migration guide now matches runtime config behavior.The auth token naming and fallback explanation are aligned with the current implementation.
Also applies to: 44-44, 51-51, 63-63
- Implement metadata merging in mem_update runtime handler\n- Complete Spanish localization for all JSON schema descriptions\n- Mark mem_session_start as unimplemented (matches runtime)\n- Relax mem_save_prompt schema constraints to handle NotImplemented state\n- Mark mem_suggest_topic_key scope as reserved/ignored\n- Generalize mem_update observation schema to allow any JSON\n- Move mem_timeline to implemented tools list in all architecture docs
There was a problem hiding this comment.
Actionable comments posted: 3
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.json (1)
46-52:⚠️ Potential issue | 🟠 MajorOutput contract still implies implemented behavior.
Line 46-Line 52 only permits
status: "started"and requiressession_id, which conflicts with the unimplemented state declared at Line 3-Line 4. This still documents a success-only response for an endpoint that returnsNotImplemented.Suggested schema adjustment
"output": { "type": "object", "properties": { "session_id": { "type": "string", "minLength": 1, "maxLength": 128 }, "status": { "type": "string", - "enum": ["started"] + "enum": ["started", "unimplemented"] } }, - "required": ["session_id", "status"], "additionalProperties": false }As per coding guidelines,
**/*: “Look for behavioral regressions, missing tests, and contract breaks across modules.”🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.json` around lines 46 - 52, The schema currently requires "status" to be "started" and mandates "session_id" which contradicts the endpoint being unimplemented; update the JSON schema in mem_session_start.json so it models both the unimplemented response and the started-success response — e.g., replace the flat enum/required with a discriminated oneOf (or an if/then) that allows {"status":"not_implemented"} (no session_id required) and {"status":"started","session_id":<string>} for success, or at minimum add "not_implemented" to the "status" enum and make "session_id" optional; adjust "status", "required" and the surrounding schema accordingly to reflect both behaviors.clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.json (1)
68-74:⚠️ Potential issue | 🟠 MajorSuccess-only output conflicts with draft/unimplemented status.
Line 68-Line 74 still documents only
status: "summarized"with requiredsession_id, but the endpoint is declared unimplemented at Line 3-Line 4. This keeps the schema contract inconsistent with runtime behavior.Suggested schema adjustment
"status": { "type": "string", - "enum": ["summarized"] + "enum": ["summarized", "unimplemented"] } }, - "required": ["session_id", "status"], "additionalProperties": falseAs per coding guidelines,
**/*: “Look for behavioral regressions, missing tests, and contract breaks across modules.”🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.json` around lines 68 - 74, The schema currently forces "status": "summarized" and requires "session_id" which conflicts with the endpoint being unimplemented/draft; update the mem_session_summary.json schema: expand the "status" enum to include the unimplemented/draft sentinel values (e.g., "unimplemented" and/or "draft") and make "session_id" optional by removing it from the "required" array (or otherwise conditionally require it if you add a JSON Schema conditional), so responses from the unimplemented endpoint validate against the contract while keeping "summarized" as a valid state.
♻️ Duplicate comments (1)
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json (1)
10-12:⚠️ Potential issue | 🟡 MinorAdd
minLength: 1tomemory_idto match runtime validation.The runtime calls
require_non_empty("memory_id", ...)attools.rs:547, but this schema allows empty strings. This was flagged in a prior review but remains unaddressed.Proposed fix
"memory_id": { "type": "string", + "minLength": 1, "description": "Identificador de memoria a actualizar." },🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json` around lines 10 - 12, The schema property "memory_id" currently allows empty strings but runtime enforces non-empty via require_non_empty (tools.rs:547); update the JSON schema for "memory_id" to add "minLength": 1 so the schema matches runtime validation and rejects empty values—locate the "memory_id" property in mem_update.json and add the minLength constraint to it.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.json`:
- Around line 30-35: The schema currently documents "candidates_count" as an
integer but the runtime actually returns an array under the key "candidates"
(runtime returns { "topic_key": slug, "candidates": [] }); update the JSON
schema to replace the "candidates_count" property with a "candidates" property
of type "array" (with appropriate item schema, e.g., strings or objects as used
at runtime) and update the "required" array to include "candidates" instead of
"candidates_count"; reference the output keys "topic_key", "candidates", and the
transform utility "extract_safe_output" when making the change so the schema
matches the tool's actual return value.
In `@modules/cerebro/src/storage/surreal.rs`:
- Around line 393-421: The timeline method currently SELECTs all rows and slices
in memory which won't scale; change the implementation of timeline to first
query the anchor row's timestamp by ID (e.g., SELECT timestamp FROM memory WHERE
memory_id = $id) to get the anchor_ts, then perform a second parameterized query
to fetch only surrounding rows using a timestamp range and limits (e.g., SELECT
... FROM memory WHERE timestamp >= $anchor_ts - ? AND timestamp <= $anchor_ts +
? ORDER BY timestamp ASC LIMIT ?, ?), applying the include_deleted filter in the
WHERE clause and mapping the query result back to Vec<MemoryRecord>; ensure you
use parameterized queries via self.db.query and proper error mapping as in the
current function.
In `@modules/cerebro/src/tools.rs`:
- Around line 573-589: The current metadata merge silently drops metadata when
record.observation is not a JSON object; update the logic around
input.input.metadata and record.observation so it fails explicitly instead of
ignoring the update: detect when record.observation is not an object (i.e.,
observation.as_object_mut() returns None) and return an Err (or propagate a
Result failure) with a clear message (e.g., "cannot merge metadata: observation
is not an object") or log a warning and return a failure; ensure the branches
handling current_metadata and observation_obj still perform the same merging
(current_obj.insert(...) and observation_obj.insert("metadata", metadata)) and
only replace the silent else branches that set *current_metadata = metadata or
skipped insertion with an explicit error path.
---
Outside diff comments:
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.json`:
- Around line 46-52: The schema currently requires "status" to be "started" and
mandates "session_id" which contradicts the endpoint being unimplemented; update
the JSON schema in mem_session_start.json so it models both the unimplemented
response and the started-success response — e.g., replace the flat enum/required
with a discriminated oneOf (or an if/then) that allows
{"status":"not_implemented"} (no session_id required) and
{"status":"started","session_id":<string>} for success, or at minimum add
"not_implemented" to the "status" enum and make "session_id" optional; adjust
"status", "required" and the surrounding schema accordingly to reflect both
behaviors.
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.json`:
- Around line 68-74: The schema currently forces "status": "summarized" and
requires "session_id" which conflicts with the endpoint being
unimplemented/draft; update the mem_session_summary.json schema: expand the
"status" enum to include the unimplemented/draft sentinel values (e.g.,
"unimplemented" and/or "draft") and make "session_id" optional by removing it
from the "required" array (or otherwise conditionally require it if you add a
JSON Schema conditional), so responses from the unimplemented endpoint validate
against the contract while keeping "summarized" as a valid state.
---
Duplicate comments:
In
`@clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json`:
- Around line 10-12: The schema property "memory_id" currently allows empty
strings but runtime enforces non-empty via require_non_empty (tools.rs:547);
update the JSON schema for "memory_id" to add "minLength": 1 so the schema
matches runtime validation and rejects empty values—locate the "memory_id"
property in mem_update.json and add the minLength constraint to it.
🪄 Autofix (Beta)
✅ Autofix completed
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 377facb7-0564-4bf8-9a9f-a5462b743091
📒 Files selected for processing (11)
clients/web/apps/docs/src/content/docs/es/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.jsonclients/web/apps/docs/src/content/docs/guides/architecture.mdmodules/cerebro/src/storage/mod.rsmodules/cerebro/src/storage/surreal.rsmodules/cerebro/src/tools.rs
📜 Review details
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (4)
- GitHub Check: sonar
- GitHub Check: pr-checks
- GitHub Check: pr-checks
- GitHub Check: Cloudflare Pages
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{md,mdx}
⚙️ CodeRabbit configuration file
**/*.{md,mdx}: Verify technical accuracy and that docs stay aligned with code changes.
For user-facing docs, check EN/ES parity or explicitly note pending translation gaps.
Files:
clients/web/apps/docs/src/content/docs/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/architecture.md
**/*
⚙️ CodeRabbit configuration file
**/*: Security first, performance second.
Validate input boundaries, auth/authz implications, and secret management.
Look for behavioral regressions, missing tests, and contract breaks across modules.
Files:
clients/web/apps/docs/src/content/docs/guides/architecture.mdmodules/cerebro/src/tools.rsclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonmodules/cerebro/src/storage/surreal.rsmodules/cerebro/src/storage/mod.rsclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_start.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.jsonclients/web/apps/docs/src/content/docs/es/guides/architecture.md
**/*.rs
⚙️ CodeRabbit configuration file
**/*.rs: Focus on Rust idioms, memory safety, and ownership/borrowing correctness.
Flag unnecessary clones, unchecked panics in production paths, and weak error context.
Prioritize unsafe blocks, FFI boundaries, concurrency races, and secret handling.
Files:
modules/cerebro/src/tools.rsmodules/cerebro/src/storage/surreal.rsmodules/cerebro/src/storage/mod.rs
🧠 Learnings (2)
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/main.rs : Preserve CLI contract unless change is intentional and documented; prefer explicit errors over silent fallback for unsupported critical paths
Applied to files:
clients/web/apps/docs/src/content/docs/guides/architecture.mdclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.json
📚 Learning: 2026-02-17T12:31:17.076Z
Learnt from: CR
Repo: dallay/corvus PR: 0
File: clients/agent-runtime/AGENTS.md:0-0
Timestamp: 2026-02-17T12:31:17.076Z
Learning: Applies to clients/agent-runtime/src/tools/**/*.rs : Implement `Tool` trait in `src/tools/` with strict parameter schema, validate and sanitize all inputs, and return structured `ToolResult` without panics in runtime path
Applied to files:
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.jsonclients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_session_summary.json
🪛 LanguageTool
clients/web/apps/docs/src/content/docs/es/guides/architecture.md
[grammar] ~45-~45: Agrega una palabra o signo.
Context: ... MCP (JSON-RPC) desde el runtime. - Herramientas Implementadas: mem_save, `mem_searc...
(QB_NEW_ES_OTHER_ERROR_IDS_MISSING_OTHER)
[grammar] ~45-~45: Corrige la mayúscula.
Context: ...C) desde el runtime. - Herramientas Implementadas: mem_save, mem_search, `mem_delet...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_ORTHOGRAPHY_UPPERCASE)
[grammar] ~45-~45: Elimina la palabra o signo.
Context: ..., mem_suggest_topic_key, mem_stats, mem_timeline. - **Herramientas Planificadas (NotImplemente...
(QB_NEW_ES_OTHER_ERROR_IDS_UNNECESSARY_OTHER)
[grammar] ~46-~46: Corrige la mayúscula.
Context: ...ts, mem_timeline. - **Herramientas Planificadas (NotImplemented)**: mem_save_prompt`, ...
(QB_NEW_ES_OTHER_ERROR_IDS_REPLACEMENT_ORTHOGRAPHY_UPPERCASE)
[grammar] ~47-~47: Elimina la puntuación
Context: ...ntext`. - Almacenamiento: Soporte multi-modelo que incluye SurrealDB embebido, JSON en...
(QB_NEW_ES_OTHER_ERROR_IDS_UNNECESSARY_PUNCTUATION)
🔇 Additional comments (10)
clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_get_observation.json (1)
10-33: LGTM — Schema now aligns with runtime behavior.The
minLength: 1constraint onmemory_idand the removal of the restrictivetype: "object"onobservationcorrectly reflect the runtime's non-empty validation and raw JSON return value. Spanish translations are in place.clients/web/apps/docs/src/content/docs/guides/architecture.md (1)
45-49: LGTM — Implemented vs planned tools now clearly documented.The tool inventory matches the runtime implementation in
modules/cerebro/src/tools.rs. This addresses the prior concern about unclear MCP endpoint status.clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_update.json (1)
33-36: Metadata merge is now implemented — documentation is accurate.The
tools.rs:573-589merge logic correctly handles themetadatafield as documented. The Spanish description "fusionar" is now valid.clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_suggest_topic_key.json (1)
15-17: LGTM —scopecorrectly marked as reserved/ignored.This addresses the prior concern about documenting behavior that doesn't exist.
modules/cerebro/src/tools.rs (1)
640-650: LGTM —mem_timelinenow calls storage layer correctly.The implementation validates inputs, checks audit permissions for
include_deleted, and delegates tostorage.timeline(). Return structure matches expected output.modules/cerebro/src/storage/mod.rs (3)
75-81: LGTM —timelinetrait method is well-defined.The signature cleanly expresses the windowing semantics. All three implementations (InMemory, DiskBacked, Surreal) use consistent filter-then-slice logic.
243-265: LGTM —InMemoryStorage::timelineimplementation is correct.Filter → sort → locate anchor → saturating window slice. The empty
Vecreturn for missingmemory_idis appropriate.
386-408: LGTM —DiskBackedStorage::timelinemirrors InMemory correctly.Identical logic ensures consistent behavior across storage backends.
clients/web/apps/docs/src/content/docs/es/guides/architecture.md (1)
43-47: LGTM — Spanish docs maintain parity with English.Tool inventory matches both the English architecture doc and runtime implementation. The capitalization in markdown emphasis (
**Herramientas Implementadas**) is intentional styling, not a grammar error.clients/web/apps/docs/src/content/docs/es/guides/cerebro/mcp-schema/mem_save_prompt.json (1)
3-4: Good alignment for unimplemented state.Marking the endpoint as draft/unimplemented and allowing
"unimplemented"inoutput.statusfixes the prior docs/runtime mismatch directionally.Also applies to: 34-34
|
Note Autofix is a beta feature. Expect some limitations and changes as we gather feedback and continue to improve it. Fixes Applied SuccessfullyFixed 3 file(s) based on 3 unresolved review comments. Files modified:
Commit: The changes have been pushed to the Time taken: |
|
Fixed 3 file(s) based on 3 unresolved review comments. Co-authored-by: CodeRabbit <noreply@coderabbit.ai>
1 participant
This pull request introduces several improvements across documentation, code formatting, and memory API schema definitions. The most significant changes are the addition of comprehensive JSON schema files for the Cerebro memory API (in Spanish documentation), updates to documentation links and structure, and code formatting cleanups for better readability.
Memory API Schema Additions:
mem_save,mem_search,mem_context,mem_get_observation,mem_delete,mem_save_prompt,mem_session_start, andmem_session_endendpoints. [1] [2] [3] [4] [5] [6] [7] [8]Documentation Improvements:
README.mdto provide clearer links to English and Spanish documentation, removed outdated instructions, and improved references to contribution guidelines.CONTRIBUTING.md, including correct paths for maintainer workflow, CI map, and security policy references. [1] [2]Code Formatting and Readability:
clients/agent-runtime/src/tools/mcp/client.rsandclients/agent-runtime/src/tools/shell.rsfor better readability, including line breaks and variable renaming for clarity. No logic changes were introduced. [1] [2] [3] [4]