PluginBench
Skill
Pass
Audit score 90

write-agentmemory-skill

rohitg00/agentmemory

Format and rules for writing consistent agentmemory skills.

What is write-agentmemory-skill?

Defines the house format and structural rules for creating or updating agentmemory skills. Use when adding a new skill, restructuring an existing one, or reviewing a skill contribution for consistency.

  • Enforces a single-tier SKILL.md format under 100 lines with required frontmatter and body sections
  • Specifies directory layout: SKILL.md (required), REFERENCE.md (optional), EXAMPLES.md (optional)
  • Requires description as two sentences: capability statement plus "Use when" triggers
  • Mandates body order: Quick start, Why, Workflow, Anti-patterns, Checklist, See also, Reference/Troubleshooting
  • Ensures generated facts (tool names, endpoints, env vars) come from source via npm run skills:gen, never hand-typed
  • Enforces style rules: no external product names, no emojis, no em-dashes, no filler

How to install write-agentmemory-skill

npx skills add https://github.com/rohitg00/agentmemory --skill write-agentmemory-skill
Claude Code
Cursor
Windsurf
Cline

How to use write-agentmemory-skill

  1. 1.Review the frontmatter requirements: name, description, optional argument-hint, and user-invocable flag
  2. 2.Write a two-sentence description: first sentence states capability, second starts with 'Use when' and lists concrete triggers
  3. 3.Structure SKILL.md body in order: Quick start example, Why (governing principle), Workflow (numbered steps with decision gates), Anti-patterns (WRONG vs RIGHT callout), Checklist, See also (one-level-deep cross-links), Reference or Troubleshooting pointer
  4. 4.Keep SKILL.md under 100 lines; move dense facts to REFERENCE.md and worked examples to EXAMPLES.md
  5. 5.Run npm run skills:gen to auto-generate tables from source code (tool names, parameters, endpoints, env vars)
  6. 6.Verify the checklist: description has "Use when" triggers, file is under 100 lines, no time-sensitive claims, concrete example present, cross-links resolve one level deep

Use cases

Good for
  • Adding a new skill to the agentmemory repository
  • Restructuring an existing skill for consistency
  • Reviewing skill contributions for format and rule compliance
  • Validating that generated facts stay in sync with source code
  • Ensuring cross-references resolve correctly and stay one level deep
Who it's for
  • Skill developers contributing to agentmemory
  • Code reviewers validating skill submissions
  • Maintainers ensuring repository consistency
  • Agents writing or updating skills programmatically

write-agentmemory-skill FAQ

What goes in the description field?

Two sentences: first states the capability in third person, second starts with 'Use when' and lists concrete triggers. Keep it under 1024 characters and distinct from sibling skills.

Should I hand-type tool names, endpoints, and env vars in SKILL.md?

No. These facts must be generated from source via npm run skills:gen. Edit the source code, then run the generator. CI will fail if generated content drifts from the code.

When should user-invocable be set to true?

Only for skills the user runs as a slash command. Reference and knowledge skills should be set to false.

Where do I put long reference material or worked examples?

Move dense facts to REFERENCE.md and worked transcripts to EXAMPLES.md. Keep SKILL.md under 100 lines and skimmable.

How deep should cross-references go?

One level deep only. Shared recovery steps belong in ../_shared/TROUBLESHOOTING.md, never inlined in SKILL.md.

Full instructions (SKILL.md)

Source of truth, from rohitg00/agentmemory.


name: write-agentmemory-skill description: The house format and rules for writing or updating an agentmemory skill. Use when adding a new skill, restructuring an existing one, or reviewing a skill contribution for consistency. user-invocable: false

agentmemory skills follow one tiered format so they stay skimmable, accurate, and current. Match it exactly.

Directory layout

plugin/skills/<name>/
  SKILL.md      (required, under 100 lines)
  REFERENCE.md  (optional, dense facts; auto-generate data tables)
  EXAMPLES.md   (optional, worked transcripts)

SKILL.md rules

  • Frontmatter: name, description, optional argument-hint, and user-invocable. Set user-invocable: true only for skills the user runs as a slash command; reference and knowledge skills are false.
  • Description is two sentences and the only thing the agent sees when deciding to load the skill. Sentence one states the capability. Sentence two starts "Use when" and lists concrete triggers. Keep it distinct from sibling skills, under 1024 chars, third person.
  • Body order: Quick start (one concrete example), Why (the governing principle), Workflow (numbered steps with decision gates), Anti-patterns (a WRONG vs RIGHT callout for the top mistake), Checklist, See also (cross-link siblings), Reference or Troubleshooting pointer.
  • Stay under 100 lines. Move dense facts to REFERENCE.md and examples to EXAMPLES.md.
  • Cross-references link one level deep only. Shared recovery steps live in ../_shared/TROUBLESHOOTING.md, never inlined.

Keep it current

Facts that exist in source (tool names and parameters, REST endpoints, env vars, connect adapters, hook events) are generated, never hand-typed. Edit the source, then run npm run skills:gen. CI runs npm run skills:check and fails on drift, so generated tables cannot fall behind the code.

Style

No external or competitor product names. No emojis. No em-dashes. No filler. State the thing and stop.

Checklist

  • Description has a "Use when" sentence with real triggers.
  • SKILL.md is under 100 lines.
  • No time-sensitive claims and no duplicated troubleshooting block.
  • Concrete example present; generated facts come from the generator.
  • Cross-links resolve and go one level deep.