Skill Claude CodeCodex
Use when explaining or changing the boundary between main Codex, worker tools, cheap gateway calls, receipts, artifacts, and deterministic gates.
2 13d ago A 32 tokens
original MIT
2Stars on the repository
16Mods indexed here, across every type
13d agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown
Skill Claude CodeCodex
Use when recreating the project environment, diagnosing local setup, running tests, or checking build and Node runtime assumptions.
2 13d ago A 28 tokens
original MIT
Skill Claude CodeCodex
Use for the executable campaign to reduce cheap-worker degradation with tiering, trajectory scoring, semantic gates, skills, and eval cases.
2 13d ago A 35 tokens
original MIT
Skill Claude CodeCodex
Use when changing worker behavior, tool contracts, routing gates, metrics, or defaults so changes stay non-blocking unless explicitly enforced.
2 13d ago A 30 tokens
original MIT
Skill Claude CodeCodex
Use when adding, changing, or auditing environment variables, tool parameters, defaults, or operator-facing configuration flags.
2 13d ago A 28 tokens
original MIT
Skill Claude CodeCodex
Use when a worker job fails, stalls, returns huge output, loses a job id, or produces a result that looks successful but is not trustworthy.
2 13d ago A 36 tokens
original MIT
Skill Claude CodeCodex
Use when measuring worker quality, cost, receipt health, artifact coverage, tool routing, or reliability-tier behavior.
2 13d ago A 29 tokens
original MIT
Skill Claude CodeCodex
Use when updating README, examples, operator docs, skill files, or external positioning copy for the worker project.
2 13d ago A 28 tokens
original MIT
Skill Claude CodeCodex
Use when comparing this worker with agent harnesses, cheap-model wrappers, benchmark claims, or external research and community ideas.
2 13d ago A 30 tokens
original MIT
Skill Claude CodeCodex
Use when a problem may have happened before and you need prior failed fixes, reverted approaches, or metrics evidence before changing code.
2 13d ago A 33 tokens
original MIT
Skill Claude CodeCodex
Use when a reliability hypothesis needs first-principles analysis, falsifiable predictions, or evidence stronger than model prose.
2 13d ago A 30 tokens
original MIT
Skill Claude CodeCodex
Use when exploring frontier ideas such as dynamic tool routing, compound tools, trajectory eval, semantic gates, context decay, or skill distillation.
2 13d ago A 34 tokens
original MIT
Skill Claude CodeCodex
Use when turning community ideas or hunches into accepted worker changes with predictions, experiments, adversarial review, and retirement criteria.
2 13d ago A 32 tokens
original MIT
Skill Claude CodeCodex
Use when operating the MCP server, running worker jobs, reading job output, inspecting artifacts, or managing long-running waits.
2 13d ago A 30 tokens
original MIT
Skill Claude CodeCodex
Use before declaring a worker change complete, accepting a worker result, or deciding whether a cheap-model output is reliable enough.
2 13d ago A 30 tokens
original MIT