- Updated docs/enforcement.md with squadChat and orchestratorDelegation gates
- Added 'For AI Agents' section with pre-flight checks, 400 error handling, and polling optimization
- Added error code reference (REVIEW_GATE_FAILED, CLOSING_COMMENT_REQUIRED, etc.)
- Added practical examples of what happens when agents violate enforcement gates
- Updated README.md with Enforcement Gates section in Feature Highlights
- Updated CHANGELOG.md with enforcement feature entry for next release
- Updated SOP-agent-task-workflow.md with enforcement gates awareness section
- All docs now reference both human operators and AI agents as primary audiences
- Add squad-post.sh for regular agent messages
- Add squad-event.sh for lifecycle events (spawned/completed/failed/status)
- Add model field to squad messages (types, server, UI, scripts)
- System events render as divider lines in squad chat panel
- Model attribution displays next to agent names in UI
- Full protocol documented in SQUAD-CHAT-PROTOCOL.md
- Updated CONTRIBUTING.md, README.md, and all SOPs
- Added VK_HOST/VK_PORT env vars to .env.example
4-check cross-model review: 10/10 (Code/GPT-5.1, Func/Grok, Perf/Grok, Sec/GPT-5.1)
Agents: R2-D2 (feature), TARS/CASE/Ava/K-2SO (reviews)
The dashboard's Success Rate, Token Usage, and Average Run Duration
graphs are powered by run.* telemetry events that agents must emit
manually — they are NOT auto-captured like task.* events.
This has broken multiple times when agents lost their AGENTS.md
instructions. Now documented in:
- AGENTS-TEMPLATE.md (copy-paste for new agents)
- SOP-agent-task-workflow.md (full API flow with telemetry steps)
Both docs now include the exact curl commands for run.started,
run.completed, and run.tokens events, plus a table clarifying
what's auto-captured vs. manual.
