doidor

doidor/agentrig

Agent

You are the developer. Implement the smallest correct change that fully satisfies the task.

doidor/agentrig

Agent

You are the judge. Score the completed work against .agentrig/eval/RUBRIC.md.

doidor/agentrig

Agent

You are the reviewer, running a different model than the developer on purpose.

doidor/agentrig

Agent

You are the security-reviewer, running a different model than the developer on purpose. You audit the diff for security, privacy, and compliance risk. You are read-only — report findings and a verdict; you do not edit code.

doidor/agentrig

Agent

You are the triager. Turn a freshly ingested task into a well-formed queued one.

doidor/agentrig

Skill Claude CodeCodex

Read PR review feedback, fix or reject each requested change, reply per thread, resolve every thread, and verify zero threads remain.

doidor/agentrig

Skill Claude CodeCodex

Diagnose and fix a failing CI run for the current branch, then re-verify.

doidor/agentrig

Skill Claude CodeCodex

Evaluate THIS repository's agent harness — a deterministic structure audit (A1) plus content quality probes (A2), plus an isolated producer/judge dynamic eval (B) with paired sign-test A/B variant comparison.

doidor/agentrig

Skill Claude CodeCodex

Record a newly-discovered gotcha to .agents/wiki/ BEFORE handoff — the harness's feedback loop. The wiki is how the next agent doesn't repeat your discovery.

doidor/agentrig

Skill Claude CodeCodex

Never cast an approving review on a PR your own (shared) agent identity authored — approval must come from an independent reviewer.

doidor/agentrig

Skill Claude CodeCodex

Rebase the PR branch onto the default branch, resolve conflicts keeping both intents, and verify mergeability before pushing.

doidor/agentrig

Skill Claude CodeCodex

Run the project's own build/test/lint and converge before handing work to a reviewer. Requires explicit baseline → after evidence — the suite must be shown to change state, not just be "green at the end".

doidor/agentrig

Skill Claude CodeCodex

Admission bar and structure for writing a new skill, so the skill library stays lean and discoverable.

doidor/agentrig

Skill Claude CodeCodex

Turn a reviewer/judge failure into an instruction-surface change that passes a prevention test.

doidor/agentrig

Skill Claude CodeCodex

General wait → inspect → fix (max 3) → self-park loop for any post-action verification (build, tests, CI, visual, lint).

doidor/agentrig

Cursor rule Cursor

What an AI reviewer should and should not flag. Keeps review high-signal.

doidor/agentrig

Cursor rule Cursor

Baseline coding standards applied to every change in this repo.

doidor/agentrig

Cursor rule Cursor

No stray debug output or debugger statements in committed code.

doidor/agentrig

Cursor rule Cursor

Security reflexes applied to every change. Specialized — highest priority.

doidor/agentrig

Instructions file GitHub Copilot

Instructions for doidor/agentrig, covering copilot instructions, critical rules (read first, every time), what this repository is, how to build, test, and lint and directory map.

doidor/agentrig

Instructions file GitHub Copilot

Instructions for doidor/agentrig: When reviewing a diff, keep signal high. Do NOT.

doidor/agentrig

Instructions file GitHub Copilot

Instructions for doidor/agentrig, a project described as: An agentic meta-harness — a harness of harnesses.

doidor/agentrig

Instructions file GitHub Copilot

Instructions for doidor/agentrig, a project described as: An agentic meta-harness — a harness of harnesses.

doidor/agentrig

Instructions file GitHub Copilot

Instructions for doidor/agentrig: Apply on every edit. When in doubt, stop and flag rather than guess.