Research-Stack/.opencode
allaun 2cbba82406 feat(agent): break-glass MCP agent for novel problem attacks
Last-resort multi-model panel via OpenRouter Fusion. Only invoked
when all other approaches (local model, single frontier, existing
skills) have failed on the same problem.

Architecture:
- Problem classification (lean_proof, code_bug, architecture, math_novel, infra_debug)
- Per-type model panel selection (Opus + DeepSeek + Gemini for proofs)
- Context assembly (files, build logs, AGENTS.md constraints)
- OpenRouter Fusion call (parallel panel + judge synthesis)
- Structured output (diagnosis, solution, alternatives, risks, cost)

Safety:
- Explicit 'break glass:' trigger required
- Cost gate: /usr/bin/bash.30-.50 per invocation, .00/session cap
- Rate limit: 3 invocations per session
- Audit trail via ContextStream

Files:
- 4-Infrastructure/shim/break_glass_mcp.py (MCP server)
- 6-Documentation/docs/specs/break_glass_mcp_agent.md (design spec)
- .opencode/skills/break-glass/SKILL.md (skill definition)
- .mcp.json (server registration)
2026-06-22 15:33:16 -05:00
..
agents docs: sync all editor rules, opencode agents, and sorry audit 2026-05-28 17:53:54 -05:00
skills feat(agent): break-glass MCP agent for novel problem attacks 2026-06-22 15:33:16 -05:00