format-skill — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited format-skill (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Transform an anonymized experience into a properly formatted, PR-ready skill file. Generates either a complete SKILL.md (for new skills) or an AMENDMENT.yaml (for amendments to existing skills).
This is the final local step before the developer submits a PR.
Generate a complete SKILL.md:
IMPORTANT: Skills MUST follow the Agent Skills Open Standard. Only name and description are required in frontmatter. Everything else goes into metadata.
---
name: [auto-generated-kebab-case]
description: [one-line description, max 1024 chars — what it does + when to use it]
license: MIT
metadata:
version: "1.0.0"
tags: [auto-detected from context]
session_tokens: [estimated tokens consumed — ask the user or read from Claude Code usage stats]
author: anonymous-contributor
source: session-contribution
---
# [Human Readable Name]
## Purpose
[Why this skill matters — what problem it prevents]
## Triggers
[When this skill should activate — specific conditions]
- [trigger 1]
- [trigger 2]
## Behavior
[Step by step what to do]
1. [step 1]
2. [step 2]
3. [step 3]
## Errors Prevented
[Real errors this prevents, with enough context to understand]
- **[Error name or message]**
[Description of what happens, why it's hard to debug,
how long it typically takes to figure out]
## Restrictions
### Hard Boundaries
[NEVER / MUST / ALWAYS rules]
### Soft Boundaries
[Prefer / by default / when possible rules]
## Self-Check
- [ ] [Verification item 1]
- [ ] [Verification item 2]
- [ ] [Verification item 3]
## Examples
### Example 1: [Happy path]
[Before/After or Scenario/Fix]
### Example 2 (Edge Case): [Edge case description]
[Before/After or Scenario/Fix]Generate an AMENDMENT.yaml:
---
type: amendment
target_skill: [skill-name]
target_version: [current version]
source_sessions: 1
proposed_version: [bumped patch or minor]
---
# Additions
## Actions (append after step [N])
[N+1]. [new step]
[N+2]. [new step]
## Errors Prevented (append)
- **[New error name]**
[Description of the new error this covers]
## Triggers (append)
- [new trigger condition]Skill file generated successfully!
Type: [New Skill / Amendment to X]
File: skills/[name]/SKILL.md (or AMENDMENT.yaml)
Repo: LaWebcapsule/d9-skills
The submit-skill step will handle cloning, copying, and PR creation.
[View file] [Edit before submitting] [Cancel]Write the skill file to a temporary location for validation. The submit-skill step will clone LaWebcapsule/d9-skills, copy the validated file into skills/[name]/, and open a PR.
Auto-generate skill names from the experience:
deploy-, config-, env-, migration-, setup-rebuild-plugins-before-deploy not fix-pluginsAssign tags based on context:
plugins/, extensions/ → [deploy, plugins].env, config → [config, environment][platform, windows/linux/macos][database, migration][ui, css][infra, aws]~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.