Skip to content

hassan-mohiddin/freeflow / .deprecated/skill-tooling-pre-rewrite/skill-eval/evaluate-skill/fixtures/v3-legacy/accepted-bundle/evidence/candidate/inputs/skill

write-skill

Verified

Create or revise agent skills. Use when defining a skill's trigger description, active instructions, stop conditions, structure, examples, references, scripts, or readiness status.

references/agent-first-instructions.md

View on GitHub
# Agent-First Instructions

Write the instruction the executing agent needs at the moment of choice.

## Strong Rules

A useful rule names:

- the trigger or pressure;
- the required or forbidden action;
- the evidence or stop condition.

Prefer:

> If the existing eval already preserves the failure and criteria, reuse it unchanged. Do not mutate evidence to prove process order.

Over:

> Ensure the evaluation workflow remains efficient and aligned with best practices.

## Placement

Order by behavioral priority:

1. user authority and source truth;
2. destructive or irreversible stops;
3. evidence and verification requirements;
4. normal execution;
5. convenience and style.

A late caveat rarely defeats an early command. Move the real constraint to where the agent first chooses.

## Editing From Failure

- Name the failed behavior.
- Find the sentence that should have prevented it.
- Sharpen or move that sentence before adding sections.
- Keep unrelated rules stable.
- Re-test the same pressure.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall