mirror of
https://github.com/sentimony/skills.git
synced 2026-09-19 07:12:05 +08:00
ihororlovskyi
3041fc51ff
The s5 eval measured the gap. All three runs read the mitigation sentence and two of three never applied it: they quoted the Codex loss clause, weighed it against the savings, and rejected the split. The sentence was written as a reporting duty - "say so in the report" - which only activates once the split is already chosen, so a run still deciding read "loses automatic access" and stopped there. The mitigation now bounds the loss before it is weighed: a root session reads the map, learns which nested file covers the directory it is about to touch, and loses one file read rather than the content. Rejecting on the bare loss without naming what the map recovers is called out as an incomplete evaluation. No decision rule is added. The spec asks the audit to evaluate the split, and all three runs did; what was missing was half the input to that evaluation.
Agent Skills
A collection of agent skills for Claude Code, Codex and other AI coding agents.
Skills
| Skill | Skill Version | Release | Description |
|---|---|---|---|
| scope-triage | 1.0.3 | v1.10.0 | Classify request scope before design work, then route to direct implementation, a light spec, or a full design cycle. |
| plan-crafting | 1.1.2 | v1.10.0 | Turn an approved design or settled requirements into a bite-sized, TDD-oriented implementation plan. |
| vitest | 1.3.0 | v1.21.0 | Configure, write, debug, run, migrate, and audit Vitest tests for JavaScript/TypeScript projects. |
| typescript | 1.4.0 | v1.20.0 | Configure tsconfig, diagnose compiler behavior, and audit or migrate TypeScript projects. |
| web-debug | 1.3.2 | v1.10.0 | Debug and verify local web apps via Playwright. |
| echarts | 1.2.0 | v1.22.0 | Build, audit, style, debug, and optimize Apache ECharts visualizations in vanilla JS, React, or Vue. |
| dashfix | 1.2.1 | v1.12.1 | Ban typographic dashes in English text, check their form where a language's orthography requires them, audit a project, and score it 0-100. |
| negafix | 1.2.1 | v1.12.1 | Ban negative parallelism ("it's not just X, it's Y"), audit prose for it, and score it 0-100. |
| commit-all | 1.1.0 | v1.16.0 | Gather the working tree into a single commit on the current branch, on explicit /commit-all invocation only. |
| maintaining-agent-context | 1.3.0 | v1.24.0 | Audit, restructure, and maintain a repository's agent instruction architecture for Claude Code and Codex. |
| frontend-crafting | 1.3.0 | v1.23.0 | Create, redesign, review, and polish user interfaces with subject-driven design decisions and a verifiable quality gate. |
Install
# All at once
npx skills add sentimony/skills -a codex claude-code -y
# Or each separately
npx skills add sentimony/skills -s scope-triage -a codex claude-code -y
npx skills add sentimony/skills -s plan-crafting -a codex claude-code -y
npx skills add sentimony/skills -s vitest -a codex claude-code -y
npx skills add sentimony/skills -s typescript -a codex claude-code -y
npx skills add sentimony/skills -s web-debug -a codex claude-code -y
npx skills add sentimony/skills -s echarts -a codex claude-code -y
npx skills add sentimony/skills -s dashfix -a codex claude-code -y
npx skills add sentimony/skills -s negafix -a codex claude-code -y
npx skills add sentimony/skills -s commit-all -a codex claude-code -y
npx skills add sentimony/skills -s maintaining-agent-context -a codex claude-code -y
npx skills add sentimony/skills -s frontend-crafting -a codex claude-code -y
Have fun ;)