Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsSkill Writing Standard — Fix Skills That Fire Inconsistently
A skeleton, a checker and an audit for your SKILL.md files: one job, no shared triggers, code for anything that must come out the same.
$7
Skill Writing Standard — Fix Skills That Fire Inconsistently
Example session with this skill installed
Audit my skills folder located at ./agent-skills and tell me which ones are failing the house standard.
- Read your context and instructions
- Compiled the writing standard
SKILL AUDIT — 2026-10-11
Pass: 9 of 14 | Over 200 lines: weekly-report | Reference files over 100 lines without contents: none
Fix first: 1) daily-digest: output format described in prose, no template 2) lead-scorer: score math done in the model's head, no script 3) weekly-report: 260 lines, split extra steps into a reference file
Shared triggers: "audit this": site-review, honest-check | "where did we leave off": session-close, daily-digest
Checked: checker run on all 14 SKILL.md files, descriptions read side by side | Not checked: scheduled jobs that call these skills
Connects securely to your tools. The creator never sees your data.
About this skill
Your skills work Monday and do something else Thursday. It's almost never length. It's two skills claiming the same trigger phrase, output formats described in prose ("something like a report card"), and counting or scoring the model does in its head. This skill fixes all three. It writes new skills in a fixed skeleton (WHAT, WHY, HOW, BOUNDARIES, TEST) with a fill-in output template, rewrites flaky ones so deterministic steps run as embedded scripts, and audits a whole skills folder with an included Python checker: PASS/FAIL per skill plus every trigger phrase two skills share. Built from an audit of a real 40-skill library. Also covers the traps nobody writes down: scripts that pass on a made-up sample and break on the real file, a "save once" script that never receives later fixes, three backticks inside an embedded script that silently end the code block, and why you retire a skill with a tombstone instead of deleting it. Python 3 for the checker; no keys, no accounts.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 8 days ago
- Passed all security checks, Safe to install
Requires
Python