skill-provenance AGENTS.md

Project instructions for Skill Provenance, a package that records the files, versions, and checksums of an AI-agent skill. A checksum is a fingerprint used to detect file changes.

In plain words
What is it for?
Use them when changing the skill bundle, updating its manifest or history, running verification, or creating its distribution packages.
Why use it?
They help agents verify that a skill bundle is intact and package or validate changes correctly.

Instructions file for CodexOpenCode

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/snapsynapse/skill-provenance/agents-md
Clone the repo
git clone --depth 1 https://github.com/snapsynapse/skill-provenance

Made for: Codex, OpenCode.

Per session 1,120 This file is loaded in full into every session.
When invoked 1,120 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.01120 $0.01120
Opus 5 $0.00560 $0.00560
Sonnet 5 $0.00224 $0.00224
Haiku 4.5 $0.00112 $0.00112

Measured 3d ago against content hash e13120288b0d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

skill-provenance AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agents Guide

Instructions for AI agents (Claude, Codex, Gemini CLI, etc.) working on this repository. This metaskill is also published to ClawHub for discovery by OpenClaw agents.

Repository structure

skill-provenance/                ← Canonical source bundle (DO NOT rename)
├── SKILL.md                     ← Agent-facing skill definition
├── README.md                    ← Human-facing user guide (has internal version header)
├── MANIFEST.yaml                ← File inventory: roles, versions, SHA-256 hashes
├── CHANGELOG.md                 ← Recent in-bundle change history (last 5 entries)
├── evals.json                   ← 41 core evaluation scenarios
├── evals-distribution.json      ← 18 supplemental distribution evals
├── validate.sh                  ← Bash script for local hash verification
├── package.sh                   ← Bash script for derived strict/ClawHub outputs
└── references/standalone-verification.md ← No-plugin verify/bootstrap path
action.yml                       ← GitHub Actions Marketplace wrapper for validate.sh
verify.sh                        ← Standalone wrapper pinned to canonical validate.sh
CHANGELOG.md                     ← Full append-only repo history
skill-provenance.skill           ← Claude Settings ZIP (rebuilt from the directory)

The skill-provenance/ directory is the single source of truth. The .skill file is a packaging wrapper.

Before making changes

  1. Read skill-provenance/MANIFEST.yaml to understand the current bundle state.
  2. Run ./skill-provenance/validate.sh to confirm all hashes are clean.
  3. Read skill-provenance/CHANGELOG.md for recent context and root CHANGELOG.md if you need older release history.

After making changes

  1. Bump per-file version integers in MANIFEST.yaml for every file you changed.
  2. Recompute hashes: run ./skill-provenance/validate.sh --update.
  3. Bump bundle_version (semver) in MANIFEST.yaml:
    • PATCH for docs, typos, non-functional changes.
    • MINOR for new features, new files, capability additions.
    • MAJOR for breaking changes to the versioning model or spec.
  4. Update bundle_date to today.
  5. Add a new entry at the top of both changelogs:
    • skill-provenance/CHANGELOG.md keeps the newest 5 entries only.
    • CHANGELOG.md at repo root is the full append-only archive. Name every file that changed and describe what changed in each.
  6. Rebuild the .skill ZIP if bundle contents changed:
    rm -f skill-provenance.skill
    zip -r skill-provenance.skill skill-provenance/
    
  7. Run ./.github/scripts/release-surface-check.sh to confirm eval-count declarations, the standalone validator pin, GuideCheck sidecar metadata, and the .skill ZIP all match the current source.

Read the full file on GitHub · 97 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 97 lines · 1,120 tokens per session scan A e13120288b0d

Subscribe to this mod's changes

skill-provenance AGENTS.md is an instructions file published in the GitHub repository snapsynapse/skill-provenance (7 stars, last pushed 5d ago), licensed MIT. It adds 1,120 tokens to every session, about $0.0056 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.