agentleFS
modeled-information-formatmodeled-information-format/mnemonic/skills/format/SKILL.md

format

Skill23 starsChanged 10 months agoMITView on GitHub

MIF (Modeled Information Format) 1.4.2 Level 3 memory templates and formatting guidelines

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 5

    1. clice-io/clice★ 1.4ksame name
    2. platformplatform/PlatformPlatform★ 439same name
    3. spiriMirror/libuipc★ 351same name
    4. WeAIClub/resume-skills★ 4same name
    5. manastalukdar/ai-devstudio★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository9 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run /mnemonic:list --namespaces to see available namespaces from loaded ontologies.

    line 18
  2. Distinctive:

    Only 4 fields are required.

    line 39
  3. Distinctive:

    That's it: Add optional fields only when you need them.

    line 54
  4. Distinctive:

    If a file with the same slug already exists, content is merged

    line 66
  5. Very specific:

    Do not use [[uuid]] wiki-links for relationships.

    line 184

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md
<!-- BEGIN MNEMONIC PROTOCOL -->

Memory

Search first: /mnemonic:search {relevant_keywords} Capture after: /mnemonic:capture {namespace} "{title}"

Run /mnemonic:list --namespaces to see available namespaces from loaded ontologies.

<!-- END MNEMONIC PROTOCOL -->

Mnemonic Format Skill

MIF (Modeled Information Format) 1.4.2 Level 3 templates and formatting guidelines.

Concept frontmatter keys are camelCase (MIF 1.4.2 section 3.3). Older mnemonic memories may still carry snake_case keys (valid_from, source_type, compressed_at), relates_to relationship types, bare-UUID targets and [[uuid]] wiki-links. Those are still read, but always write the forms below. tools/mnemonic-migrate-mif upgrades old files on request.

Trigger Phrases

  • "memory format", "MIF format", "frontmatter template", "memory template", "mnemonic schema"

Quick Start: Minimal Memory

Only 4 fields are required. Everything else is optional.

---
id: 550e8400-e29b-41d4-a716-446655440000
title: "Human-readable title"
type: semantic
created: 2026-01-23T10:30:00Z
---

# Title

Content here.

That's it. Add optional fields only when you need them.


File Naming

{slug}.memory.md
  • slug: URL-safe title, lowercase, max 50 chars
  • UUID is stored only in the frontmatter id: field
  • If a file with the same slug already exists, content is merged
  • MIF 1.4.2 names concept files {slug}.md and calls the .memory infix legacy. mnemonic keeps {slug}.memory.md because its search, hooks and tools all glob *.memory.md.

Full Optional Fields Reference

---
# REQUIRED (always include these 4)
id: 550e8400-e29b-41d4-a716-446655440000
title: "Human-readable title"
type: semantic
created: 2026-01-23T10:30:00Z

# COMMON
modified: 2026-01-23T14:22:00Z
namespace: _semantic/decisions/project
tags:
  - architecture
  - database

# BI-TEMPORAL TRACKING
temporal:
  validFrom: 2026-01-23T00:00:00Z
  validUntil: null
  recordedAt: 2026-01-23T10:30:00Z
  ttl: P90D
  decay:
    model: exponential
    halfLife: P7D
    strength: 0.85
  accessCount: 5
  lastAccessed: 2026-01-23T14:22:00Z

# PROVENANCE
provenance:
  sourceType: agent_inferred      # user_explicit | user_implicit | agent_inferred | external_import | system_generated
  sourceRef: file:///path/to/source.ts:42
  agent: claude-opus-4
  confidence: 0.95
  trustLevel: moderate_confidence # verified | user_stated | high_confidence | moderate_confidence | low_confidence | uncertain
  sessionId: abc123

# RELATIONSHIPS (authoritative; mirrored in the body, see below)
relationships:
  - type: relates-to
    target: urn:mif:7c9e6679-7425-40de-944b-e07fc1f90ae7
  - type: supersedes
    target: urn:mif:f47ac10b-58cc-4372-a567-0e02b2c3d479
    metadata:
      label: "Replaced after the 2026 review"

# CODE REFERENCES (mnemonic extension)
codeRefs:
  - file: src/auth/handler.ts
    line: 42
    symbol: authenticateUser
    type: function                # function | class | method | variable | type

# CITATIONS (MIF 1.4.2 section 5.4)
citations:
  - "@type": Citation
    citationType: documentation   # article | book | paper | website | documentation | repository | video | podcast | specification | dataset | tool | other
    citationRole: supports        # see MIF 1.4.2 section 5.4.4
    title: "Source Title"
    url: https://example.com/source
    author: "Author Name"
    date: 2026-01-23
    accessed: 2026-01-23
    relevance: 0.90
    note: "Brief description"

# COMPRESSION (auto-generated by gc --compress)
summary: "Concise 2-3 sentence summary (max 500 chars)"
compressedAt: 2026-01-24T10:00:00Z
---

# Memory Title

Content in markdown format.

## Relationships

- relates-to [Other Memory Title](urn:mif:7c9e6679-7425-40de-944b-e07fc1f90ae7)
- supersedes [Old Memory Title](urn:mif:f47ac10b-58cc-4372-a567-0e02b2c3d479)

Cognitive Memory Types

TypeUse ForNamespace Pattern
semanticFacts, specs, configs, definitions_semantic/{subtype}/project
episodicDebug sessions, incidents, discoveries_episodic/{subtype}/project
proceduralRunbooks, how-to guides, workflows_procedural/{subtype}/project

Relationship Syntax

Relationships are authoritative in the frontmatter relationships array and every entry is mirrored as a Markdown link in a ## Relationships section at the end of the body (MIF 1.4.2 section 5.3), one per line:

## Relationships

- {type} [{Target Title}](urn:mif:{target-uuid})
  • {type} is a kebab-case token (relates-to, not relates_to or RelatesTo).
  • The target is urn:mif:{uuid} of the target memory, the same string in frontmatter and body.
  • An optional human note goes in metadata.label, not a top-level label.
  • Do not use [[uuid]] wiki-links for relationships; they are legacy.
TypeMeaning
relates-toGeneral association
supersedesReplaces older memory (inverse: superseded-by)
derived-fromBuilt upon another memory (inverse: derives)
conflicts-withConflicts with another memory
part-ofComponent of a larger whole (inverse: contains)
implementsRealizes a design/decision (inverse: implemented-by)
usesUtilizes a technology or tool (inverse: used-by)

Quick Templates

Generate REAL values for ids and timestamps. Run uuidgen | tr '[:upper:]' '[:lower:]' and date -u +"%Y-%m-%dT%H:%M:%SZ" before creating any memory. NEVER leave placeholders.

See Quick Start above for the minimal template. Below are additional templates for common memory types.

Decision Memory

---
id: <generate with: uuidgen | tr '[:upper:]' '[:lower:]'>
type: semantic
namespace: _semantic/decisions/project
created: <generate with: date -u +"%Y-%m-%dT%H:%M:%SZ">
title: "Decision: {what}"
tags: [architecture]
provenance:
  sourceType: user_explicit
  confidence: 0.95
---

# Decision: {What}

## Quick Answer
{What} was chosen for {primary reason}.

## Context
- **Alternatives:** {Option A}, {Option B}
- **Key drivers:** {Driver 1}, {Driver 2}
- **Trade-offs:** {What was sacrificed}

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.