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.
npx agentmods add instructions/wei18/upkeep/claude-mdgit clone --depth 1 https://github.com/wei18/UpkeepWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01113 | $0.01113 |
| Opus 5 | $0.00557 | $0.00557 |
| Sonnet 5 | $0.00223 | $0.00223 |
| Haiku 4.5 | $0.00111 | $0.00111 |
Grade A, and why
Upkeep CLAUDE.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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Upkeep (repo-audit-action)
Local dir is repo-audit-action/; the published name is wei18/upkeep — the mismatch is intentional. Full reference: docs/en/design.md.
What this is
An AI repo auditor that catches drift (stale docs, spec/code mismatch, orphaned assets, convention violations). Pipeline: discovery → parallel reviewers → synthesis → HTML report. Three entry points, one engine:
- CI: reusable workflow
.github/workflows/audit.yml(on: workflow_call) + composite actions under.github/actions/. Runs on the caller's checkout; Upkeep's code arrives viauses: wei18/upkeep/.github/actions/<x>@<tag>. Each reviewer is an independent matrix job. Upserts one tracking issue. - Marketplace: root
action.yml, a composite action for the conventional- uses: wei18/upkeep@<tag>step syntax and a GitHub Marketplace listing. Same engine and inputs as the reusable workflow, but runs as a single job so reviewers execute sequentially instead of in parallel. - Local:
scripts/local-audit.sh <target>— same flow,claude -psubprocesses, temp-dir intermediates via--add-dir. Never creates issues; prints the summary instead.skills/upkeep-audit/SKILL.mdis a thin wrapper (clones engine to~/.cache/upkeep, runs script, summarizes). The CI workflow does not route through the skill.
Hard invariants
- Output-only: never edit or delete anything in the audited repo. No exceptions.
- No presumed ground truth: report divergence with evidence; don't assume code > docs or vice versa.
- Workflow interface (
with:inputs, secrets, permissions) is a public contract — changes need a major version.
Plugin / skill distribution
Three install paths, all pointing at skills/upkeep-audit/ (do not move it):
- Claude Code plugin:
.claude-plugin/marketplace.json(marketplaceupkeep) → pluginupkeep, single-skill layout (SKILL.md at plugin root) → invocation/upkeep:upkeep-audit. The plugin version is pinned in two files that must stay in lockstep — bump both on skill changes:skills/upkeep-audit/.claude-plugin/plugin.jsonand theplugins[].versionfield in.claude-plugin/marketplace.json. npx skills add wei18/upkeep --skill upkeep-audit— relies on the flatskills/<name>/SKILL.mdlayout andname+descriptionfrontmatter. Don't break either.- Manual copy into
~/.claude/skills/.
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.
- 2d ago First seen · 49 lines · 1,113 tokens per session scan A dfa974fc7969
Upkeep CLAUDE.md is an instructions file published in the GitHub repository wei18/Upkeep (14 stars, last pushed 5d ago), licensed MIT. It adds 1,113 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 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.