cerebrocybersolutionscerebrocybersolutions/libro/skills/memory-attribution-lint/SKILL.md

memory-attribution-lint

Skill1 starsChanged 39 days agoApache-2.0View on GitHub

Validate and auto-inject the 11-key provenance frontmatter contract on memory files at write time. Use when authoring memory entries, running lint passes on memory directories, or backfilling missing attribution fields. Enforces a uniform provenance schema (name, description, type, source_surface, host, tool, session_id, written_at_utc, source_path, confidence, lifecycle) so every memory entry is traceable to its origin surface, tool, and session.

In this repository12 other files it gives its agents.
- advisor-modeSkill
- brain-setupSkill
- ceo-briefSkill
- cerebro-doctorSkill
- council-modeSkill
- daily-briefSkill
- decisionSkill
- memory-recallSkill
- orchestrator-modeSkill
- sessionendSkill
- sessionstartSkill
- skill-manageSkill

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Very specific:
Write-time validator + auto-injector for memory file frontmatter.
line 14

≈ “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

memory-attribution-lint

Write-time validator + auto-injector for memory file frontmatter. Enforces an 11-key provenance contract on .md files in memory directories. Closes the "where did this memory come from?" gap by capturing source surface, tool, host, and session at write time.

Why provenance matters

Memory entries accumulate across sessions, tools, and surfaces. Without provenance, stale or contradictory entries are hard to triage. The 11-key contract gives every memory file a complete audit trail: who wrote it, from what tool, on what host, in what session, at what time, with what confidence, and at what lifecycle stage.

Contract

Required frontmatter keys:

Key	Source	Auto-derived
name	author	NO
description	author	NO
type	author (user/feedback/project/reference)	NO
source_surface	env CLAUDE_SURFACE or default	YES
host	`socket.gethostname()`	YES
tool	env CLAUDE_TOOL or default `unknown`	YES
session_id	env CLAUDE_SESSION_ID or `unknown`	YES
written_at_utc	`datetime.now(timezone.utc).isoformat()`	YES
source_path	author declares	NO
confidence	author: high/med/low	NO
lifecycle	author: pinned/active/archived/closed/superseded/stale-by:YYYY-MM-DD	NO

Scripts

scripts/lint.py: validate + auto-inject; exit 0=pass, 1=blocked, 2=injected
scripts/backfill.py: bulk backfill existing memory files
scripts/tests/fixtures/: golden samples for validation cases

Invocation

# Validate single file
python3 scripts/lint.py path/to/memory.md

# Dry-run (report only)
python3 scripts/lint.py --dry-run path/to/memory.md

# Backfill all memory dirs
python3 scripts/backfill.py --dry-run
python3 scripts/backfill.py --execute

Hook wire (PreToolUse)

See .claude/settings.json hooks.PreToolUse → Write/Edit matching **/memory/*.md. The hook fires lint on any Write/Edit to a file matching the pattern; exit 1 blocks the write until missing required keys are supplied, exit 2 auto-injects derivable keys and allows the write to proceed.

Environment variables consumed

Variable	Purpose	Default if unset
`CLAUDE_SURFACE`	Tool surface label (cc / hermes / etc.)	`unknown`
`CLAUDE_TOOL`	Specific tool that initiated the write	`unknown`
`CLAUDE_SESSION_ID`	Session identifier for cross-reference	`unknown`

Set these at session start to capture full provenance automatically. Without them, the required keys still pass validation (defaulted to unknown) but lose audit value.

Scope Contract (Least-Privilege)

Dimension	Scope
Read paths	Any `.md` file passed as target argument
Write paths	Same file (auto-inject mode only); never writes outside target
MCP / tool surface	None — pure Python, stdlib only
Network egress	None
Surface	Any (hook fires at write-time regardless of surface)
Credentials	None
Escalation trigger	Exit 1 (blocked write) — operator must add missing required keys manually

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.

Check your setup, free