scriptoriumadmin/claude-skills
1
0
Fork 0
mirror of https://github.com/alirezarezvani/claude-skills.git synced 2026-10-11 03:37:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
claude-skills/engineering/agent-harness
History
Claude 8232162630
fix(productivity): resolve deep-work asset paths + lift meetings agentic signals 
- check_paths.py: fix two unresolvable shutdown_checklist.md references
  in cs-deep-work agent/command (correct ../skills/ prefix)
- meetings SKILL.md: add genuine intake/iteration/close-out discipline
  (clarify-the-decision intake, re-run-until-fit stop condition,
  owner+date completion check) -> manifest signals now 5/5, checklist
  still PASS at 97 lines
- productivity harness manifest regenerated

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQLKzYb1bR2LYqwYUupm5f
2026-07-17 06:17:54 +00:00
..
.claude-plugin feat(engineering): agent-harness skill + agentic-readiness audit of both engineering domains 2026-07-03 06:01:43 +00:00
agents feat(engineering): agent-harness skill + agentic-readiness audit of both engineering domains 2026-07-03 06:01:43 +00:00
commands feat(engineering): agent-harness skill + agentic-readiness audit of both engineering domains 2026-07-03 06:01:43 +00:00
skills/agent-harness fix(productivity): resolve deep-work asset paths + lift meetings agentic signals 2026-07-17 06:17:54 +00:00
README.md feat(engineering): agent-harness skill + agentic-readiness audit of both engineering domains 2026-07-03 06:01:43 +00:00

README.md

agent-harness

Turn any domain folder of this repo into a bounded agentic loop: pick up a goal, compile it into tasks with machine-run verification, execute, verify, retry with caps, escalate to a human when budgets exhaust, and close only when everything is verified.

GOAL → goal_compiler → PLAN → loop_controller: [execute → verify]* → CLOSE
                                     ↑ retry ≤ caps, changed approach
                                     └ ESCALATE — never fake success

What ships

Piece Purpose
scripts/harness_manifest_builder.py Scan a domain folder → manifest.v1 JSON (skills, tools, checks, agentic signals)
scripts/goal_compiler.py Goal + manifest → plan.v1 task plan; refuses vague goals (exit 3, forcing questions)
scripts/loop_controller.py init/next/record/verify/close/status state machine; controller runs checks itself
assets/harnesses/*.json 18 committed per-domain manifests (regenerable, diff-stable)
assets/harness_manifest.schema.json Manifest schema
references/ Agentic-loop canon, verification discipline, domain-harness design (cited)
agents/harness-runner.md Stateless one-task-per-invocation executor
commands/cs-harness.md /cs:harness <domain> <goal> end-to-end driver

All tools are stdlib-only, pass --help and --sample, and emit JSON.

Design lineage

Anthropic's long-running-agents harness (feature-list + stateless shifts), verifier's law, SWE-agent's environment-feedback lesson, Ralph-loop fresh-context iteration, Cognition's serialize-writers rule, and this repo's own tc-tracker / autoresearch locked-evaluator / loop-library stop-state primitives. See skills/agent-harness/references/.