Borrowing it
Nothing to install: this file belongs to scalefocus/skilly. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/scalefocus/skilly/main/CLAUDE.mdgit clone --depth 1 https://github.com/scalefocus/skillyWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/scalefocus/skilly/claude-md)<a href="https://agentmods.dev/instructions/scalefocus/skilly/claude-md"><img src="https://agentmods.dev/badge/instructions/scalefocus/skilly/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/scalefocus/skilly/claude-md"><img src="https://agentmods.dev/badge/instructions/scalefocus/skilly/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.03749 | $0.03749 |
| Opus 5.5 | $0.01500 | $0.01500 |
| Sonnet 5.5 | $0.00750 | $0.00750 |
| Haiku 4.5 | $0.00375 | $0.00375 |
Grade A, and why
skilly 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 15d 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 — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — skilly project context
Read this first. It pins the project's intent, decisions, conventions, and the single most important constraint. The authoritative spec is
SKILLY_SPEC.md; this file is the working context for day-to-day implementation.§nreferences throughout this file are the canonical deep-dive pointers intoSKILLY_SPEC.md.
What this is
skilly — an enterprise-grade, open-source, self-hosted registry for governing
Anthropic-style SKILL.md agent skills across an organization and its business units,
with identity/access anchored in Microsoft Entra ID. Greenfield, SKILL.md-compatible.
Apache 2.0.
Key files (jump straight here)
| File | Role |
|---|---|
SKILLY_SPEC.md |
The authoritative spec — every change lands here first |
packages/shared/src/version.ts |
APP_VERSION — bumped on every app change |
packages/web/src/app/whats-new/changelog.ts |
The What's new CHANGELOG — one entry per version bump |
packages/shared/src/external-tool.ts |
The pinned install wire format — the ONLY place it may change |
packages/web/e2e/shots.mjs |
Playwright script that captures Quick-start / manual screenshots (manual built ad-hoc, outside the repo) |
db/migrations/ |
Plain-SQL migrations, applied in order by the migrate compose service |
deploy/docker-compose.yml |
The full stack (postgres, migrate, minio, clamav, web, worker, proxy) |
The change workflow (GATED — spec-first with a mandatory stop)
Every change to app behavior goes through this gate. Never skip straight to code.
- Grill first. Before touching anything, interrogate the request rigorously: ask pointed clarifying questions — scope, edge cases, roles affected, visibility implications, data model impact, what happens on expiry/deletion/conflict — until the requirement is unambiguous. Depth is proportional to the change (a copy fix needs one confirming question; a new feature needs a real interview). Do not assume answers you could ask for.
- Update
SKILLY_SPEC.md— and ONLY the spec. Write the behavior, data, and contracts into the relevant sections. No code, no migrations, no manual edits yet. - STOP for verification. Present the spec diff (what changed, where, and the decisions it encodes) and end the turn. The user must review and approve the spec before anything is built. Do not proceed on silence; wait for explicit go-ahead. If the user amends the spec, loop back to step 1/2 as needed.
- Implement to match the approved spec. Code follows spec, never the reverse. New code ships with its tests in the same change: unit tests for any new domain/RBAC/semver/validation logic, integration tests for anything touching the API, DB, or SCIM surface, and e2e coverage when the change alters a user-facing flow (propose→review→publish→install and the like). Untested new code is an incomplete implementation, not a follow-up.
- Release ritual. Version bump → changelog → commit (see the "Before every commit" checklist below).
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.
- 15d ago Changed · +4 lines · +71 tokens per session 4573b50e268e
- 17d ago Changed 0d1b2988171f
- 25d ago Changed · +7 lines · +135 tokens per session 8291bb65827e
- 1mo ago Changed · +61 tokens per session 5f219837d03c
- 1mo ago Changed · +3 lines · +69 tokens per session 8e083625c952
- 2mo ago First seen · 188 lines · 3,413 tokens per session scan A 3e14448a99cf
skilly CLAUDE.md is an instructions file published in the GitHub repository scalefocus/skilly (11 stars, last pushed yesterday), licensed Apache-2.0. It adds 3,749 tokens to every session, about $0.0150 per session on Opus 5.5. 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.
skilly CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| skilly CLAUDE.mdscalefocus/skilly | Instructions file | 11 | yesterday | Apache-2.0 | A | 3,749 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |