canvas-org/meta-agent
Continual harness optimization
Continual harness optimization
Bring your own agent and build a self-improving agentic system. Automatically mine failures, optimize the agent harness, and gate against regressions.
A meta-skill that designs domain-specific agent teams, defines specialized agents, and generates the skills they use.
Turn ad-hoc Claude Code / Codex / Cursor sessions into a repeatable loop — research → plan → wave-execute → close. Parallel sub-agents, inter-wave quality gates, dual GitHub+GitLab integration, F2 Memory + Persona-Panel + gsd-Pattern Adoption. 37 skills · 18 commands · 13 agents · 7360 tests.
Engineering decisions engine that know when they're stale. Frame, compare, decide — with evidence decay and parity enforcement. For Claude Code, Cursor, Gemini CLI, Codex and more.
autonomous harness engineering
2 captures since 2026-05-23
AI agent config detected
Key config paths
examples/python_cli_benchmark/baseline/AGENTS.md
examples/python_cli_benchmark/baseline/GEMINI.md
examples/python_fixture_benchmark/baseline/AGENTS.md
examples/python_fixture_benchmark/baseline/GEMINI.md