feat(coordinator): complete track 4 slice 1 foundations#607
Conversation
…onality and update timeout durations
# Conflicts: # clients/rook/Cargo.lock # clients/rook/Cargo.toml # clients/rook/migrations/0001_initial.sql # clients/rook/src/db/account.rs # clients/rook/src/db/mod.rs # clients/rook/src/db/pool.rs # clients/rook/src/db/route.rs # clients/rook/src/services/account.rs # clients/rook/src/services/health.rs # clients/rook/src/services/pool.rs # clients/rook/src/services/route.rs
Deploying corvus with
|
| Latest commit: |
0ef9b53
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://ceba98bc.corvus-42x.pages.dev |
| Branch Preview URL: | https://feat-track-4-slice-1-coordin.corvus-42x.pages.dev |
|
Warning Rate limit exceeded
Your organization is not enrolled in usage-based pricing. Contact your admin to enable usage-based pricing to continue reviews beyond the rate limit, or try again in 12 minutes and 30 seconds. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (8)
📝 WalkthroughWalkthroughThis PR introduces a new in-process multi-agent coordinator module for Corvus that supervises parallel child agents with deterministic state transitions, envelope-based messaging, and fan-in aggregation. The coordinator is integrated into the delegate tool and accompanied by comprehensive specifications and timeout standardization across provider/channel configs. Changes
Sequence Diagram(s)sequenceDiagram
actor Parent as Parent Agent
participant C as Coordinator
participant DAR as DelegatedAgentRunner
participant Child as Child Agent
participant R as Registry
Parent->>C: run(CoordinatorLaunchRequest)
C->>C: transition(Dispatching)
Note over C: admit_child() for each child
C->>R: register ChildRecord
C->>C: transition(Supervising)
par Child Execution
C->>DAR: spawn_child_task
DAR->>Child: configure & execute
Child->>DAR: CodeSessionResult
DAR->>C: MessageEnvelope<ChildCompleted>
C->>R: apply_envelope, update state
and Supervision Loop
C->>C: monitor JoinSet for completion
alt First child fails
C->>C: transition(Cancelling)
C->>Child: cancel_token
else Parent cancelled
C->>Child: cancel_token
C->>C: transition(Cancelling)
else All succeed
C->>C: transition(Completed)
end
end
C->>Parent: CoordinatorOutcome
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes The coordinator module itself is dense with state machine logic, concurrency primitives (mutex-protected registries, atomic sequence counter, CancellationToken), and envelope sequencing/validation that demands careful correctness review. The delegate tool integration adds another layer of concern (outcome mapping, fail-closed semantics). While the duration refactorings are largely homogeneous and low-risk, the overall heterogeneity and logic density (particularly around state transitions, child supervision, and envelope correlation) warrant thorough analysis for race conditions, edge cases, and behavior correctness. Possibly related PRs
Suggested labels
Suggested reviewers
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
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: 2025-04-20 to 2026-04-20 |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 10
🤖 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/agent-runtime/src/agent/coordinator.rs`:
- Around line 1346-1362: The test
coordinator_slice_defers_non_in_process_transport_and_deferred_scope is brittle
because it uses include_str!("coordinator.rs") and substring checks; replace
this string-based check with a compile-time type-level assertion or
exhaustiveness check against the enum CoordinatorTransport so CI fails only when
the actual type/variants change. For example, remove the include_str! usage and
instead use a static assertion (e.g., static_assertions::assert_type_ne_all! or
a const/compile-time match) that only compiles if CoordinatorTransport contains
only the allowed InProcess variant and does not contain RemoteBridge,
CrossProcess, MailboxPersistence, or WorktreeIsolation; keep the test name but
change its body to perform the compile-time guard against the unwanted variants
rather than string inspections.
- Around line 403-410: The match on error.downcast_ref::<AgentExecutionError>()
in the coordinator logic currently only lists the two known variants
(AgentExecutionError::IterationBudgetExceeded and ::CostBudgetExceeded) — add an
explicit catch-all Some(_) arm that returns CodeSessionStatus::Error so new
variants map deterministically to Error instead of causing future compile/merge
issues; update the match near where AgentExecutionError and CodeSessionStatus
are used to include Some(_) => CodeSessionStatus::Error while leaving None =>
CodeSessionStatus::Error unchanged.
In `@clients/agent-runtime/src/observability/log.rs`:
- Line 472: Replace the invalid call to Duration::from_mins(2) with a valid
std::time::Duration construction, e.g. Duration::from_secs(120); locate
occurrences of the Duration::from_mins(...) call (such as the duration:
Duration::from_mins(2) initializer in
clients/agent-runtime/src/observability/log.rs) and update each to use
Duration::from_secs(number_of_minutes * 60) (or an equivalent from_secs value)
across the codebase.
In `@clients/agent-runtime/src/tools/delegate.rs`:
- Around line 1162-1214: The test name and assertions overclaim because
execute() only consumes "agent", "prompt", and "context" so the runtime never
sees the deferred fields; either rename
session_mode_preserves_fail_closed_boundaries_for_deferred_transport_and_escalation
to something like session_mode_schema_rejects_deferred_transport_fields to
reflect that it only validates the schema, or add a new non-readonly test that
constructs the same DelegateTool (using DelegateTool::new with
DelegateExecutionMode::Session) but sets AutonomyLevel::Supervised and injects a
StubSessionCoordinatorExecutor (or similar test double used by your session
coordinator path) so the stub can assert that the coordinator request produced
by execute() does not contain transport, mailbox, remote_bridge, worktree, or
permission_escalation; ensure the new test calls execute() and inspects the
coordinator call arguments rather than relying solely on the schema assertions.
In `@clients/web/apps/dashboard/src/composables/chatOnboardingContract.spec.ts`:
- Line 15: Remove the unstable import of tmp/CLAUDIO_ROADMAP.md from
chatOnboardingContract.spec.ts (the import of claudioRoadmap) — either commit
that file to the repo or replace the import with a committed fixture/inline test
string so the test can resolve in CI; then update the assertions that currently
match exact prose (the assertions around the roadmap verification in the spec)
to assert on stable markers instead (e.g., headings, IDs, comment anchors, or a
simple regex/contains for a heading like "##" or "Roadmap" rather than verbatim
lines) so future rewording of the roadmap won’t break the test.
In `@clients/web/apps/dashboard/vite.config.js`:
- Around line 7-10: The Vite dev server currently sets server.fs.allow:
[repoRoot], exposing the whole repository; narrow this to only the necessary
paths (e.g., the openspec, tmp, and clients/composeApp directories referenced by
tests) or make the broad allowlist conditional for test mode only. Update the
vite.config.js server.fs.allow configuration (the server.fs.allow array) to list
explicit absolute paths derived from repoRoot joined with the specific
directories used by the dashboard contract tests (e.g., path.join(repoRoot,
"openspec"), path.join(repoRoot, "tmp"), path.join(repoRoot,
"clients/composeApp")) or add a guard around this setting that uses a TEST_MODE
flag to permit repoRoot only during tests. Ensure you modify the server.fs.allow
entry and remove the single repoRoot entry so high-risk source directories are
not exposed.
In `@clients/web/apps/dashboard/vite.config.ts`:
- Around line 6-14: The Vite config currently exposes the whole repo by setting
server.fs.allow: [repoRoot]; change it so that broad repo access is enabled only
in test mode and otherwise only the dashboard root and the specific test file
are allowed: detect test mode via process.env.NODE_ENV or a dedicated flag
(where defineConfig and repoRoot are referenced), and update server.fs.allow to
conditionally include either repoRoot (for tests) or a narrowed array containing
the dashboard root path (compute from fileURLToPath(new URL(...,
import.meta.url)) and the explicit tmp/CLAUDIO_ROADMAP.md path). Apply the same
conditional change to the second allow block referenced around the defineConfig
section at lines ~34-38 (the other server.fs.allow usage) so only tests get
repoRoot while normal dev gets the minimal allowed paths.
In
`@openspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/verify-report.md`:
- Line 1: The first markdown heading "## Verification Report" is level 2; change
it to a top-level heading by replacing it with a level 1 heading "# Verification
Report" so the file starts with an H1 and satisfies markdownlint.
In `@openspec/specs/multi-agent-orchestration/spec.md`:
- Around line 61-85: The spec and the code diverge: EnvelopeMeta (symbols:
EnvelopeMeta, coordinator_id, child_id, sequence, correlation_id, sent_at,
transport) conveys sender/recipient implicitly and message kind is the
CoordinatorMessage enum tag in payload; add a short non-normative note to the
spec clarifying that "kind is conveyed by the typed payload variant
(CoordinatorMessage) and sender/recipient are derived from coordinator_id +
child_id plus message direction" so readers can map the spec to the Rust struct,
or alternatively update EnvelopeMeta to include explicit sender/recipient/kind
fields to make the mapping explicit (choose one approach and apply
consistently).
🪄 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: 2a5ccac6-8d1e-4a2c-9f03-381aab987584
📒 Files selected for processing (51)
clients/agent-runtime/src/agent/classifier.rsclients/agent-runtime/src/agent/coordinator.rsclients/agent-runtime/src/agent/mod.rsclients/agent-runtime/src/agent/unified_loop.rsclients/agent-runtime/src/channels/cli.rsclients/agent-runtime/src/channels/irc.rsclients/agent-runtime/src/channels/lark.rsclients/agent-runtime/src/channels/telegram.rsclients/agent-runtime/src/channels/whatsapp.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/daemon/mod.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/gateway/cerebro.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/identity.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/observability/log.rsclients/agent-runtime/src/providers/anthropic.rsclients/agent-runtime/src/providers/compatible.rsclients/agent-runtime/src/providers/copilot.rsclients/agent-runtime/src/providers/gemini.rsclients/agent-runtime/src/providers/glm.rsclients/agent-runtime/src/providers/mod.rsclients/agent-runtime/src/providers/ollama.rsclients/agent-runtime/src/providers/openai.rsclients/agent-runtime/src/providers/openai_codex.rsclients/agent-runtime/src/providers/openrouter.rsclients/agent-runtime/src/providers/pool.rsclients/agent-runtime/src/security/policy.rsclients/agent-runtime/src/skillforge/integrate.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/tools/composio.rsclients/agent-runtime/src/tools/delegate.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/transcription/whisper_cli.rsclients/agent-runtime/src/update/mod.rsclients/agent-runtime/tests/admin_config_api_integration.rsclients/web/apps/dashboard/src/composables/chatOnboardingContract.spec.tsclients/web/apps/dashboard/vite.config.jsclients/web/apps/dashboard/vite.config.tsopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/design.mdopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/exploration.mdopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/proposal.mdopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/specs/agent-loop/spec.mdopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/specs/multi-agent-orchestration/spec.mdopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/state.yamlopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/tasks.mdopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/verify-report.mdopenspec/specs/agent-loop/spec.mdopenspec/specs/multi-agent-orchestration/spec.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). (1)
- GitHub Check: Cloudflare Pages
🧰 Additional context used
📓 Path-based instructions (11)
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/skillforge/integrate.rsclients/agent-runtime/src/providers/anthropic.rsclients/agent-runtime/src/providers/openrouter.rsclients/agent-runtime/src/agent/classifier.rsclients/agent-runtime/src/providers/pool.rsclients/agent-runtime/src/observability/log.rsclients/agent-runtime/src/tools/composio.rsclients/agent-runtime/src/providers/ollama.rsclients/agent-runtime/src/daemon/mod.rsclients/agent-runtime/src/providers/gemini.rsclients/agent-runtime/src/transcription/whisper_cli.rsclients/agent-runtime/src/identity.rsclients/agent-runtime/src/gateway/cerebro.rsclients/agent-runtime/src/providers/compatible.rsclients/agent-runtime/src/security/policy.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/update/mod.rsclients/agent-runtime/src/providers/glm.rsclients/agent-runtime/src/channels/lark.rsclients/agent-runtime/src/providers/openai.rsclients/agent-runtime/src/channels/telegram.rsclients/agent-runtime/src/channels/cli.rsclients/agent-runtime/src/agent/unified_loop.rsclients/agent-runtime/src/providers/mod.rsclients/agent-runtime/src/channels/irc.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/providers/openai_codex.rsclients/agent-runtime/src/providers/copilot.rsclients/agent-runtime/src/agent/mod.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/channels/whatsapp.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/tools/delegate.rsclients/agent-runtime/src/agent/coordinator.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/skillforge/integrate.rsclients/agent-runtime/src/providers/anthropic.rsclients/agent-runtime/src/providers/openrouter.rsclients/agent-runtime/src/agent/classifier.rsclients/agent-runtime/src/providers/pool.rsclients/agent-runtime/src/observability/log.rsclients/agent-runtime/src/tools/composio.rsclients/agent-runtime/src/providers/ollama.rsclients/agent-runtime/src/daemon/mod.rsclients/agent-runtime/src/providers/gemini.rsclients/agent-runtime/src/transcription/whisper_cli.rsclients/agent-runtime/src/identity.rsclients/agent-runtime/tests/admin_config_api_integration.rsclients/agent-runtime/src/gateway/cerebro.rsclients/agent-runtime/src/providers/compatible.rsclients/agent-runtime/src/security/policy.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/update/mod.rsclients/agent-runtime/src/providers/glm.rsclients/agent-runtime/src/channels/lark.rsclients/agent-runtime/src/providers/openai.rsclients/agent-runtime/src/channels/telegram.rsclients/agent-runtime/src/channels/cli.rsclients/agent-runtime/src/agent/unified_loop.rsclients/agent-runtime/src/providers/mod.rsclients/agent-runtime/src/channels/irc.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/providers/openai_codex.rsclients/agent-runtime/src/providers/copilot.rsclients/agent-runtime/src/agent/mod.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/channels/whatsapp.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/tools/delegate.rsclients/agent-runtime/src/agent/coordinator.rs
**/*.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:
clients/agent-runtime/src/skillforge/integrate.rsclients/agent-runtime/src/providers/anthropic.rsclients/agent-runtime/src/providers/openrouter.rsclients/agent-runtime/src/agent/classifier.rsclients/agent-runtime/src/providers/pool.rsclients/agent-runtime/src/observability/log.rsclients/agent-runtime/src/tools/composio.rsclients/agent-runtime/src/providers/ollama.rsclients/agent-runtime/src/daemon/mod.rsclients/agent-runtime/src/providers/gemini.rsclients/agent-runtime/src/transcription/whisper_cli.rsclients/agent-runtime/src/identity.rsclients/agent-runtime/tests/admin_config_api_integration.rsclients/agent-runtime/src/gateway/cerebro.rsclients/agent-runtime/src/providers/compatible.rsclients/agent-runtime/src/security/policy.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/update/mod.rsclients/agent-runtime/src/providers/glm.rsclients/agent-runtime/src/channels/lark.rsclients/agent-runtime/src/providers/openai.rsclients/agent-runtime/src/channels/telegram.rsclients/agent-runtime/src/channels/cli.rsclients/agent-runtime/src/agent/unified_loop.rsclients/agent-runtime/src/providers/mod.rsclients/agent-runtime/src/channels/irc.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/providers/openai_codex.rsclients/agent-runtime/src/providers/copilot.rsclients/agent-runtime/src/agent/mod.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/channels/whatsapp.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/tools/delegate.rsclients/agent-runtime/src/agent/coordinator.rs
**/*
⚙️ 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/agent-runtime/src/skillforge/integrate.rsclients/agent-runtime/src/providers/anthropic.rsclients/agent-runtime/src/providers/openrouter.rsclients/agent-runtime/src/agent/classifier.rsclients/agent-runtime/src/providers/pool.rsclients/agent-runtime/src/observability/log.rsclients/agent-runtime/src/tools/composio.rsclients/agent-runtime/src/providers/ollama.rsclients/agent-runtime/src/daemon/mod.rsclients/agent-runtime/src/providers/gemini.rsclients/agent-runtime/src/transcription/whisper_cli.rsclients/agent-runtime/src/identity.rsclients/agent-runtime/tests/admin_config_api_integration.rsclients/agent-runtime/src/gateway/cerebro.rsclients/agent-runtime/src/providers/compatible.rsclients/agent-runtime/src/security/policy.rsclients/web/apps/dashboard/vite.config.tsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/update/mod.rsclients/agent-runtime/src/providers/glm.rsclients/agent-runtime/src/channels/lark.rsclients/web/apps/dashboard/vite.config.jsclients/agent-runtime/src/providers/openai.rsclients/agent-runtime/src/channels/telegram.rsclients/agent-runtime/src/channels/cli.rsclients/agent-runtime/src/agent/unified_loop.rsclients/agent-runtime/src/providers/mod.rsopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/tasks.mdclients/agent-runtime/src/channels/irc.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/providers/openai_codex.rsopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/verify-report.mdclients/agent-runtime/src/providers/copilot.rsclients/web/apps/dashboard/src/composables/chatOnboardingContract.spec.tsclients/agent-runtime/src/agent/mod.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/channels/whatsapp.rsclients/agent-runtime/src/config/schema.rsopenspec/specs/multi-agent-orchestration/spec.mdclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/tools/delegate.rsclients/agent-runtime/src/agent/coordinator.rsopenspec/specs/agent-loop/spec.md
clients/agent-runtime/src/providers/**/*.rs
📄 CodeRabbit inference engine (clients/agent-runtime/AGENTS.md)
Implement
Providertrait insrc/providers/and register insrc/providers/mod.rsfactory when adding a new provider
Files:
clients/agent-runtime/src/providers/anthropic.rsclients/agent-runtime/src/providers/openrouter.rsclients/agent-runtime/src/providers/pool.rsclients/agent-runtime/src/providers/ollama.rsclients/agent-runtime/src/providers/gemini.rsclients/agent-runtime/src/providers/compatible.rsclients/agent-runtime/src/providers/glm.rsclients/agent-runtime/src/providers/openai.rsclients/agent-runtime/src/providers/mod.rsclients/agent-runtime/src/providers/openai_codex.rsclients/agent-runtime/src/providers/copilot.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/composio.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/tools/delegate.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/composio.rsclients/agent-runtime/src/gateway/cerebro.rsclients/agent-runtime/src/security/policy.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/tools/delegate.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/composio.rsclients/agent-runtime/src/gateway/cerebro.rsclients/agent-runtime/src/security/policy.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/tools/delegate.rs
clients/agent-runtime/src/channels/**/*.rs
📄 CodeRabbit inference engine (clients/agent-runtime/AGENTS.md)
Implement
Channeltrait insrc/channels/with consistentsend,listen, andhealth_checksemantics and cover auth/allowlist/health behavior with tests
Files:
clients/agent-runtime/src/channels/lark.rsclients/agent-runtime/src/channels/telegram.rsclients/agent-runtime/src/channels/cli.rsclients/agent-runtime/src/channels/irc.rsclients/agent-runtime/src/channels/whatsapp.rs
**/*.{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:
openspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/tasks.mdopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/verify-report.mdopenspec/specs/multi-agent-orchestration/spec.mdopenspec/specs/agent-loop/spec.md
clients/agent-runtime/src/main.rs
📄 CodeRabbit inference engine (clients/agent-runtime/AGENTS.md)
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
Keep startup path lean and avoid heavy initialization in command parsing flow
Files:
clients/agent-runtime/src/main.rs
🧠 Learnings (13)
📚 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:
clients/agent-runtime/src/skillforge/integrate.rsclients/agent-runtime/src/tools/composio.rsclients/agent-runtime/src/daemon/mod.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/channels/cli.rsopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/tasks.mdclients/agent-runtime/src/providers/openai_codex.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/agent/coordinator.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 : Preserve CLI contract unless change is intentional and documented; prefer explicit errors over silent fallback for unsupported critical paths
Applied to files:
clients/agent-runtime/src/skillforge/integrate.rsclients/agent-runtime/src/providers/anthropic.rsclients/agent-runtime/src/agent/classifier.rsclients/agent-runtime/src/tools/composio.rsclients/agent-runtime/src/daemon/mod.rsclients/agent-runtime/src/providers/gemini.rsclients/agent-runtime/src/transcription/whisper_cli.rsclients/agent-runtime/src/identity.rsclients/agent-runtime/tests/admin_config_api_integration.rsclients/agent-runtime/src/gateway/cerebro.rsclients/agent-runtime/src/providers/compatible.rsclients/agent-runtime/src/security/policy.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/update/mod.rsclients/agent-runtime/src/providers/glm.rsclients/agent-runtime/src/channels/lark.rsclients/agent-runtime/src/channels/cli.rsclients/agent-runtime/src/agent/unified_loop.rsclients/agent-runtime/src/providers/mod.rsclients/agent-runtime/src/channels/irc.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/providers/openai_codex.rsclients/agent-runtime/src/providers/copilot.rsclients/agent-runtime/src/agent/mod.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/tools/delegate.rsclients/agent-runtime/src/agent/coordinator.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/providers/**/*.rs : Implement `Provider` trait in `src/providers/` and register in `src/providers/mod.rs` factory when adding a new provider
Applied to files:
clients/agent-runtime/src/providers/openrouter.rsclients/agent-runtime/src/providers/compatible.rsclients/agent-runtime/src/providers/glm.rsclients/agent-runtime/src/agent/mod.rsclients/agent-runtime/src/agent/coordinator.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/composio.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/tools/delegate.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/**/Cargo.toml : Do not add heavy dependencies for minor convenience; justify new crate additions
Applied to files:
clients/agent-runtime/src/tools/composio.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/channels/cli.rsclients/agent-runtime/src/agent/mod.rsclients/agent-runtime/src/agent/coordinator.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:
clients/agent-runtime/src/daemon/mod.rsclients/agent-runtime/src/identity.rsclients/agent-runtime/src/security/policy.rsclients/agent-runtime/src/providers/mod.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/config/schema.rsclients/agent-runtime/src/gateway/admin.rsclients/agent-runtime/src/tools/delegate.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/**/*.rs : Avoid unnecessary allocations, clones, and blocking operations to maintain performance and efficiency
Applied to files:
clients/agent-runtime/src/identity.rsclients/agent-runtime/src/skills/mod.rsclients/agent-runtime/src/providers/mod.rsclients/agent-runtime/src/providers/openai_codex.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/agent/coordinator.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/skills/mod.rsclients/agent-runtime/src/gateway/cost.rsclients/agent-runtime/src/tools/shell.rsclients/agent-runtime/src/main.rsclients/agent-runtime/src/agent/coordinator.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/channels/**/*.rs : Implement `Channel` trait in `src/channels/` with consistent `send`, `listen`, and `health_check` semantics and cover auth/allowlist/health behavior with tests
Applied to files:
clients/agent-runtime/src/channels/telegram.rsclients/agent-runtime/src/gateway/mod.rsclients/agent-runtime/src/channels/whatsapp.rsclients/agent-runtime/src/agent/coordinator.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/**/Cargo.toml : Preserve release-size profile assumptions in `Cargo.toml` and avoid adding heavy dependencies unless clearly justified
Applied to files:
clients/agent-runtime/src/channels/cli.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/**/*.rs : Never log secrets, tokens, raw credentials, or sensitive payloads in any logging statements
Applied to files:
clients/agent-runtime/src/providers/mod.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:
openspec/specs/multi-agent-orchestration/spec.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/{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:
clients/agent-runtime/src/agent/coordinator.rs
🪛 markdownlint-cli2 (0.22.0)
openspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/verify-report.md
[warning] 1-1: First line in a file should be a top-level heading
(MD041, first-line-heading, first-line-h1)
🔇 Additional comments (47)
clients/agent-runtime/src/channels/cli.rs (1)
469-469: No issues in this change.Formatting-only cleanup in
format!keeps output and behavior intact.clients/agent-runtime/src/skills/mod.rs (3)
634-634: No regression in discover output formatting.This is a safe formatting-only cleanup; output and behavior stay unchanged.
1284-1284: Local-skill update skip path remains correct.
bail!semantics and message intent are preserved; this change is non-functional.
1375-1375: Missing official subdir error handling remains intact.This is a syntax/format cleanup only; failure behavior is unchanged and still explicit.
clients/agent-runtime/src/skillforge/integrate.rs (1)
169-169: LGTM: Formatting cleanup.Trailing comma removed from
format!arguments. No behavioral change.clients/agent-runtime/src/update/mod.rs (1)
1584-1584: No issue —Duration::from_minsis stable.
Duration::from_minswas stabilized in Rust 1.91.0 and requires no feature flags on stable Rust. The API is already widely used throughout the codebase without issues. No action needed.> Likely an incorrect or invalid review comment.clients/agent-runtime/src/channels/irc.rs (1)
15-15: No issue.std::time::Duration::from_minswas stabilized in Rust 1.91.0, and the project uses toolchain 1.92. The API is available on stable and the code is valid.clients/agent-runtime/src/channels/telegram.rs (1)
2227-2227: MSRV requirement met: workspace uses Rust 1.92.The workspace is configured with
rust-toolchainchannel 1.92, which satisfies the requirement forDuration::from_mins(stabilized in Rust 1.91.0). No build issues will occur.clients/agent-runtime/src/identity.rs (1)
457-457: LGTM: Idiomatic Rust refactoring for deterministic ordering.The change from
.sort_by()to.sort_by_key()is a clean improvement that makes the sorting logic more idiomatic and potentially more efficient. This aligns well with the PR's focus on deterministic behavior for multi-agent coordination, and the existing test at line 565 confirms the ordering remains correct.Also applies to: 475-475
clients/agent-runtime/src/transcription/whisper_cli.rs (1)
395-395: No issues found.Duration::from_minsis a validstd::time::Durationconstructor (available since Rust 1.53.0). The change fromDuration::from_secs(120)toDuration::from_mins(2)is correct and semantically equivalent; no action required.clients/agent-runtime/src/channels/lark.rs (3)
128-128: LGTM — clearer heartbeat duration.The timeout remains 5 minutes and still exceeds the default 120-second ping interval.
309-310: LGTM — fragment GC cutoff stays bounded.This keeps the stale-fragment retention at 5 minutes with more readable duration construction.
413-413: LGTM — dedup retention is unchanged and clearer.The 30-minute dedup window is preserved while avoiding the manual seconds expression.
clients/agent-runtime/src/gateway/cerebro.rs (1)
863-879: Same MSRV check applies toDuration::from_mins.The test fixture still uses a 5-minute idempotency TTL, so behavior is unchanged. Please confirm the Rust toolchain/MSRV supports
Duration::from_mins; otherwise keepDuration::from_secs(300).clients/agent-runtime/src/gateway/admin.rs (1)
2543-2565: Same MSRV check applies to this test fixture TTL.
Duration::from_mins(5)preserves the previous 300-second window. Verify the repo’s Rust version supports it; otherwise useDuration::from_secs(300).clients/agent-runtime/src/gateway/cost.rs (1)
342-389: Same MSRV check applies to both cost gateway test states.Both helpers keep the same 5-minute idempotency TTL. Confirm the project toolchain supports
Duration::from_mins; otherwise useDuration::from_secs(300).clients/agent-runtime/tests/admin_config_api_integration.rs (1)
71-87: Same MSRV check applies to the 1-minute integration-test TTL.The idempotency duration remains 60 seconds. Verify
Duration::from_minsis supported by the declared Rust toolchain; otherwise useDuration::from_secs(60).clients/agent-runtime/src/security/policy.rs (1)
109-122: No action required—Duration::from_hours()is compatible with your toolchain.Rust 1.92 (declared in
rust-toolchain.toml) supports the convenience constructorsfrom_hours()andfrom_mins(), both stabilized in Rust 1.84.0. The codebase already uses these consistently across 40+ locations (channels, gateways, providers, security). The one-hour window is preserved, and the code is semantically sound.clients/agent-runtime/src/gateway/mod.rs (1)
3678-3678: No action needed. The project'srust-toolchain.tomldeclares Rust 1.92, which exceeds the 1.91.0 minimum required forDuration::from_mins(). The code change is compatible.clients/agent-runtime/src/channels/whatsapp.rs (1)
353-353: No action needed —Duration::from_hoursis stable.
std::time::Duration::from_hours()was stabilized in Rust 1.91.0 (October 2025) and is a valid stable API as of now (April 2026). The code compiles correctly and requires no changes.> Likely an incorrect or invalid review comment.clients/agent-runtime/src/providers/glm.rs (1)
93-95: Equivalent timeout normalization.This keeps the same 2-minute request timeout and leaves the connect timeout unchanged.
clients/agent-runtime/src/providers/anthropic.rs (1)
181-183: Equivalent timeout normalization.This preserves the same overall request timeout while keeping connection setup bounded separately.
clients/agent-runtime/src/tools/composio.rs (1)
40-42: Equivalent timeout normalization.This keeps the Composio client at a 1-minute request timeout without broadening network behavior.
clients/agent-runtime/src/providers/openrouter.rs (1)
116-118: Equivalent timeout normalization.This keeps the OpenRouter request timeout unchanged at 2 minutes.
clients/agent-runtime/src/providers/gemini.rs (1)
228-230: Equivalent timeout normalization.This preserves the existing 2-minute Gemini request timeout.
clients/agent-runtime/src/providers/pool.rs (1)
516-519: Equivalent test fixture update.The cooldown remains one minute in the future and still exercises account skipping deterministically.
clients/agent-runtime/src/providers/ollama.rs (1)
79-81: Equivalent timeout normalization.This keeps Ollama’s longer 5-minute request timeout intact for local/remote model calls.
clients/agent-runtime/src/providers/openai.rs (1)
141-143: No MSRV issue.Duration::from_minsis available since Rust 1.53.0; the repository is pinned to Rust 1.92, which fully supports this API. The timeout change is safe.> Likely an incorrect or invalid review comment.clients/agent-runtime/src/daemon/mod.rs (1)
497-503: LGTM — minute-based expectations are equivalent.The configured and floored updater interval assertions still match the production calculation.
clients/agent-runtime/src/main.rs (2)
2224-2228: LGTM — callback wait remains three minutes.This is a readability-only timeout constructor change with no OAuth flow behavior change.
2890-2894: LGTM — the threshold assertion remains equivalent.The override still validates a 90-minute staged-image reaper threshold.
clients/agent-runtime/src/providers/compatible.rs (1)
58-62: LGTM — provider timeout value is unchanged.Both constructors still use a two-minute request timeout and preserve the existing connect timeout/fallback behavior.
Also applies to: 80-84
clients/agent-runtime/src/providers/copilot.rs (2)
249-253: LGTM — Copilot request timeout remains two minutes.The HTTP client behavior is unchanged.
1019-1021: LGTM — the polling-delay expectation is clearer.The test still asserts the 5-second interval plus 3-second safety margin.
clients/agent-runtime/src/providers/mod.rs (1)
237-250: Loop refactor is clean and behavior-preserving.Switching to
while letimproves clarity without changing scan/replace semantics, including the bare-prefix continuation path.clients/agent-runtime/src/config/schema.rs (2)
3326-3341: LGTM.The catalog URL validation still fails closed for non-HTTPS and loopback targets.
3650-3694: LGTM.Treating a whitespace-only Cerebro auth token as missing closes the configuration gap when an endpoint is set.
clients/agent-runtime/src/providers/openai_codex.rs (2)
90-94: LGTM.The overall request timeout remains two minutes and the connect timeout is unchanged.
279-294: LGTM.The
while letloop keeps the same SSE frame processing and trailing-buffer behavior.clients/agent-runtime/src/agent/mod.rs (1)
5-5: LGTM.Exporting
coordinatorfromagentmatches the new orchestration module boundary.clients/agent-runtime/src/agent/classifier.rs (1)
59-60: LGTM.
Reverse(rule.priority)preserves the descending-priority rule order with clearer intent.clients/agent-runtime/src/tools/shell.rs (1)
60-60: SameDuration::from_minsMSRV concern already raised onunified_loop.rs.openspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/tasks.md (1)
5-30: Archive checklist flip looks consistent with the rest of the PR.Every referenced implementation file in tasks 1.1–5.2 has a matching diff (
coordinator.rs,delegate.rs,CLAUDIO_ROADMAP.md, specs). No discrepancies.openspec/specs/agent-loop/spec.md (1)
491-519: Requirement scoping is tight and matches implementation.The "MUST NOT be interpreted as enabling remote child transport, disk-backed mailbox delivery, worktree isolation, or full delegated permission escalation" line mirrors the deferred-scope assertions enforced by
coordinator.rs::coordinator_slice_defers_non_in_process_transport_and_deferred_scope— good traceability.clients/agent-runtime/src/tools/delegate.rs (1)
47-77: Clean seam; fail-closed fallbacks read well.The
SessionCoordinatorExecutortrait cleanly isolates the coordinator from the tool layer for testability, andsession_result_from_outcome/fail_closed_session_resultcorrectly fall back to a non-successToolResultin every branch where the child slot is empty (children.first()== None). Preserves the existingToolResultcontract and keeps the OneShot path untouched.Also applies to: 149-204
clients/agent-runtime/src/agent/coordinator.rs (1)
480-510: Solid lock discipline and deterministic aggregation — nice work on this slice.
std::sync::Mutexheld only around tight synchronous critical sections (no awaits), correctly converting poison toFailedClosed.JoinSet+ parent-ownedCancellationTokengives deterministic cancel/fail propagation without a bespoke message bus.ordered_outcomes()sorting bylaunch_indexguarantees the spec's "stable aggregation regardless of completion order" requirement, and theaggregate_results_preserve_launch_ordertest actually proves it with reversed delays.- First-failure-wins guard (
!failedon Line 888) pluscoordinator_cancellation.cancel()gives the "fatal child failure cancels siblings" semantic cleanly.Ship it (after the smaller nits above).
Also applies to: 821-957
clients/agent-runtime/src/agent/unified_loop.rs (1)
16-16: MSRV is sufficient — change is valid.Rust 1.92 (workspace toolchain) supports
Duration::from_mins(stabilized in 1.81). Both changes inunified_loop.rs:16andshell.rs:60compile and run without issue.
| pub fn admit_child(&self, request: &ChildLaunchRequest) -> Result<(), CoordinatorError> { | ||
| let mut registry = self | ||
| .registry | ||
| .lock() | ||
| .map_err(|_| CoordinatorError::FailedClosed("registry lock poisoned".to_string()))?; | ||
|
|
||
| if registry.contains_key(&request.child_id) { | ||
| return Err(CoordinatorError::DuplicateChild(request.child_id.0.clone())); | ||
| } | ||
|
|
||
| if registry | ||
| .values() | ||
| .any(|record| record.launch_index == request.launch_index) | ||
| { | ||
| return Err(CoordinatorError::DuplicateLaunchIndex(request.launch_index)); | ||
| } | ||
|
|
||
| registry.insert( | ||
| request.child_id.clone(), | ||
| ChildRecord { | ||
| child_id: request.child_id.clone(), | ||
| agent_name: request.agent_name.clone(), | ||
| launch_index: request.launch_index, | ||
| session_id: None, | ||
| state: ChildState::Registered, | ||
| last_sequence: 0, | ||
| terminal_reason: None, | ||
| summary: None, | ||
| }, | ||
| ); | ||
| Ok(()) | ||
| } |
There was a problem hiding this comment.
Partial-admit state leak on duplicate child.
admit_child is called in a loop before transition(Dispatching) (Lines 836–839). If children[0] admits successfully but children[1] is a duplicate / has a colliding launch_index, the registry now contains children[0] while the coordinator is still in Initialized and the caller gets an Err. A retry on the same Coordinator instance would then either double-admit children[0] or silently skip it depending on caller logic.
Two safe fixes: (a) validate all children up-front (uniqueness of child_id and launch_index) before any registry mutation, or (b) document that a failed run/run_with_cancellation invalidates the coordinator and callers must construct a fresh one. Today's callers (only DefaultSessionCoordinatorExecutor) always build a fresh Coordinator, so this is latent rather than exploitable — but worth closing before slice 2 reuses coordinators.
Also applies to: 836-839
- coordinator.rs: swap string-based test for exhaustive compile-time match - coordinator.rs: add Some(_) catch-all for future AgentExecutionError variants - log.rs: replace invalid Duration::from_mins(2) with from_secs(120) - delegate.rs: rename test to reflect schema-only validation - chatOnboardingContract.spec.ts: inline roadmap fixture, use regex assertions - vite.config.js: narrow server.fs.allow to specific paths - vite.config.ts: make fs.allow conditional on test mode - verify-report.md: change H2 to H1 heading - spec.md: add non-normative note on implicit kind/sender/recipient
1 participant
Related Issues
Closes #525
Summary
clients/agent-runtime, including the coordinator state machine, supervised child registry, in-process messaging envelopes, fan-out/fan-in behavior, and session-mode delegate routing through the coordinator seamopenspec/specs/agent-loop/spec.mdandopenspec/specs/multi-agent-orchestration/spec.md, and keeptmp/CLAUDIO_ROADMAP.mdaligned with delivered scope and deferred Track 4 gapsTested Information
cargo fmt --manifest-path clients/agent-runtime/Cargo.toml --all -- --checkcargo check --manifest-path clients/agent-runtime/Cargo.tomlcargo clippy --manifest-path clients/agent-runtime/Cargo.toml --all-targets -- -D warningscargo test --manifest-path clients/agent-runtime/Cargo.tomlmake web-test-allpnpm --dir clients/web checkpnpm --dir clients/web/apps/dashboard exec vitest run src/composables/chatOnboardingContract.spec.ts --environment happy-domDocumentation Impact
tmp/CLAUDIO_ROADMAP.mdopenspec/specs/agent-loop/spec.mdopenspec/specs/multi-agent-orchestration/spec.mdopenspec/changes/archive/2026-04-20-track-4-slice-1-coordinator-foundations/Breaking Changes
None.
Checklist