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 skills/abilityai/cornelius/ref-auditnpx skills add Abilityai/cornelius --skill ref-auditgit clone --depth 1 https://github.com/Abilityai/corneliusWhat 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 | $0.00101 | $0.01912 |
| Opus 5 | $0.00051 | $0.00956 |
| Sonnet 5 | $0.00020 | $0.00382 |
| Haiku 4.5 | $0.00010 | $0.00191 |
Grade A, and why
ref-audit 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ref Audit
ℹ️ First, set expectations: before anything else, print one short line with this skill's version and its most recent change — the top entry of
metadata.changelogabove — e.g.ref-audit v1.0 — recent: initial version. Then proceed.
The integrity health check for a reference scope. It answers "is the source of truth still canonical, well-formed, and honestly stamped?" It is read-only: it produces a report and prints a summary. It never edits (repairs/merges are /ref-reconcile, which is human-gated) — that is exactly what makes it autonomous-safe.
Read first: resources/layered-brains/COMPANY-BRAIN-SCHEMA.md (the enum + temporal contract) and resources/layered-brains/REFERENCE-SCOPE-SCHEMA.md (the reference-kind contract). This skill checks notes against those docs.
State Dependencies
| Source | Location | Read | Write | Description |
|---|---|---|---|---|
| Company family | Brain/Company/{people,orgs,products,engagements,market}/*.md |
✓ | Entity notes audited | |
| Reports | resources/company-brain/reports/ |
✓ | ref-audit-YYYY-MM-DD.md |
The contract this audits
- provenance: every note must be
provenance: reference. Anything else = a leak (a cognitive note crept in, or an insight was mis-filed here). - type ∈
person · organization · product · agent · engagement · interaction. - relationship ∈ (by
type) person:self-team·client-contact·counsel·prospect·partner-contact; organization:self·client·prospect·partner·vendor·competitor·investor; product:own·competitor.agent/engagement/interactioncarry norelationship. - as_of: present + a valid
YYYY-MM-DD. - status ∈
active · superseded; asupersedednote MUST have asuperseded_by: "[[...]]"that resolves. - partition rule: a
relationship: competitornote belongs inmarket/; a transacted entity does not. - links: every
[[wiki-link]]resolves to an existing note; every entity has at least one relation (no orphans).
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 · 120 lines · 101 tokens per session scan A 647e5257880d
ref-audit is a skill published in the GitHub repository Abilityai/cornelius (104 stars, last pushed 9d ago), licensed MIT. It adds 101 tokens to every session and 1,912 once invoked, about $0.0005 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…