Demo automation, docs cleanup, and embedding provider fix#5
Merged
Conversation
New Playwright script (demo-competition.spec.ts) optimized for judging criteria: Impact, Opus 4.6 Use, and Depth & Execution. 7-act story arc: hook with completed consensus, platform tour with agent persona deep-dive, live community+thread creation, real-time deliberation with emergent phase transitions, human intervention with energy spike, institutional memory knowledge graph, and closing contrast with a second completed thread. Accompanying voiceover script (VOICEOVER_SCRIPT.md) with per-act cues, adaptable narrator lines, recording tips, and key phrases mapped to each judging criterion. https://claude.ai/code/session_01QezAFNoTPMs4Q1KaQqCb4M
Cut Act 2 expertise tab and separate members view. Reduced deliberation loop from 8 to 5 iterations. Removed Act 5 continuation loop and cost summary spotlight. Shortened Act 6 graph interaction. Eliminated Act 7 enzyme engineering detour — go straight to closing shot. Compressed all pause durations. Voiceover script trimmed to match tighter timing. https://claude.ai/code/session_01QezAFNoTPMs4Q1KaQqCb4M
Shorter, phase-appropriate agent responses via prompt word limits (EXPLORE=250, DEBATE=300, DEEPEN=250, CONVERGE=150) backed by configurable per-phase max_tokens as a hard API cap. Fix content duplication in parse_agent_response by actually stripping structured sections. Render post content as markdown in the frontend using react-markdown with theme-aware component mappings. Update wiki (Agent-System, Deliberation-Engine) and add wiki TOC to README. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Reduce README from 459 to 203 lines by replacing verbose sections (API table, project structure tree, institutional memory deep-dive, demo seeding, testing categories) with concise summaries linking to the wiki. Rewrite docs/README.md as a spec index with wiki cross-references. Add wiki pointer headers to all 5 spec files. Fix stale content: Next.js -> React 19, 6 agents -> 10, add phase_max_tokens to EngineConfig and engine.yaml examples. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Relax observer thresholds and double phase token limits for faster phase progression and richer agent responses - Propagate full YAML config (energy + observer) through SessionManager instead of loading only phase_max_tokens - Persist extracted memories to DB and detect cross-references after deliberation completion - Add DB fallback in memory list and graph endpoints when in-memory store is empty - Fix seed script to handle paginated posts response format and auto- detect Docker Compose postgres - Render consensus content as markdown in ConsensusReveal - Reconnect WebSocket on session resume when deliberation is running - Update tests for new default config values Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add screenshots.spec.ts for automated wiki screenshot capture (requires headed mode for WebGL graph rendering) - Update OBSERVER_SPEC.md and SYSTEM_DESIGN.md with relaxed thresholds (hysteresis 2, window 5, doubled token limits) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Use create_embedding_provider() so memory extraction and cross-reference detection respect the EMBEDDING_PROVIDER env var instead of always using mock embeddings. Also remove unused extract_entities import. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
MockEmbeddingProviderin memory extraction and cross-reference detection to respectEMBEDDING_PROVIDERenv varTest plan
uv run ruff check .passesuv run ruff format --check .passesuv run pytest tests/ -x -q -m "not slow and not integration"-- 686 passed🤖 Generated with Claude Code