Convert skill authoring into a RED-GREEN-REFACTOR cycle with pressure scenarios.
npx skills add https://github.com/wenerme/ai --skill writing-skills-wenerme
Please help me install this Agent Skill. Skill: writing-skills Source: https://github.com/wenerme/ai/tree/main/skills/writing-skills Command: npx skills add https://github.com/wenerme/ai --skill writing-skills-wenerme
This Skill reduces the risk that newly written or edited agent skills are unclear or non-compliant by forcing a test-first loop before deployment, mirroring RED-GREEN-REFACTOR for documentation.
Use the writing-skills workflow to create failing pressure scenarios first, then iteratively update the target SKILL.md until the same scenarios reliably comply.
High-intent search queries and answers about installing and using this skill.
Test-driven documentation hardens agent skills by applying a RED-GREEN-REFACTOR cycle to authoring. You create failing pressure scenarios first, capture baseline failures with verbatim rationalizations, then iteratively update the skill text until scenarios reliably comply.
Pressure scenario testing for skill authoring uses realistic subagent scenarios to evaluate skill documentation. It captures exact rationalizations and failure modes under combined pressures, enabling targeted loophole hardening through minimal skill edits and repeated re-verification.
You write bulletproof agent skills by designing pressure scenarios that expose rationalization countermeasures. Capture baseline failures verbatim, apply minimal edits to add explicit negations and red flags to the SKILL.md, and re-verify to close loopholes.
Yes, you can verify skill compliance under multiple combined pressures simultaneously using subagents. The workflow requires capturing baseline failures, applying minimal skill edits to address observed violations, and repeating re-verification until all loopholes are closed.
The best way to refactor skill text to close documentation loopholes is by adding explicit negations, counters, and red flags. This targeted loophole hardening prevents repeat violations discovered during pressure scenario testing and improves discovery metadata.
You should not use the test-first loop for skill authoring when deploying documentation without compliance risks. If intuition-based authoring suffices and pressure scenario testing or subagent verification is unnecessary, this cycle adds overhead without proportionate benefit.
seanGSISGTest skills with subagents to verify resistance to rationalization under pressure.
ace3000chaoApply the dead-end framework to mobilize teams under extreme pressure.
venhdevApply RED-GREEN-REFACTOR cycles to author and verify agent skills.
jmdueaCreate, edit, and validate Claude skills with TDD-style guidelines and pressure-scenario templates.
