SKILL PLAN · DETERMINISTIC

A small plan for a real job

Start with the minimum useful stack. Add nothing unless it has a clear responsibility in the run.

Task
Debug and reproduce
Agent
Codex
Priority
Balanced
Start over

Rules + reviewed metadata only. No LLM. Same inputs produce the same plan.

03 SKILLS · Install order

Your Skill Plan

Install and verify one role at a time. The plan never forces three Skills when fewer are enough.

Read the public task guide ↗
01Core execution

Native · Codex

repo-intake-and-plan

Maps the relevant execution path before changing code.
Open profile and install guide ↗
Review permissions, limitations, and evidence
Agent fit
Native · Codex
Main limitation
It does not create environments, download assets, execute commands, perform paper lookup, or write the final report.
First risk
Planning from incomplete repository evidence · Medium
Source
Pinned source · 3ab50525
02Guardrail

Native · Codex

grill-me

Forces a minimal run and challenges unsupported assumptions.
Open profile and install guide ↗
Review permissions, limitations, and evidence
Agent fit
Native · Codex
Main limitation
A relentless interview can be tiring and disproportionate for small, reversible decisions.
First risk
Decision fatigue · Low
Source
Pinned source · 885e2ca4
03Evidence

Native · Codex

tdd

Turns the reproduced failure into a lasting regression check.
Open profile and install guide ↗
Review permissions, limitations, and evidence
Agent fit
Native · Codex
Main limitation
The workflow needs a functioning test runner and clear public behavior.
First risk
False confidence · Medium
Source
Pinned source · 885e2ca4

Combined permission budget

This is the union of declared permissions across the plan. Review each package before installation.

Review package instructions before installRead project contextRead filesWrite testsRun test commands

Finish with evidence

Reduce an intermittent report to the smallest repeatable failure and a regression check.

Observe first; change one variable at a time and preserve the original failure evidence.

  1. 3-step acceptance check
  2. Reproduce the same failure twice from a documented starting state.
  3. Identify the smallest input or state transition that triggers it.
  4. Add a check that fails before the fix and passes after it.
  • Stop conditions
  • The reproduction depends on unknown external state.
  • Multiple variables are changed before the failure is isolated.
  • Logs may contain secrets or personal data.