scribe CLAUDE.md

scribe CLAUDE.md is an instructions file for coding agents from Naoray/scribe. It costs 1,743 tokens per session, scanned A, original, MIT.

Repository instructions for Scribe, a command-line tool for managing local AI coding-agent skills. They describe its JSON response format, error codes, command discovery, and automation guidance.

In plain words
What is it for?
Use them to list or manage coding-agent skills, inspect command schemas, automate Scribe commands, and handle their results and errors.
Why use it?
They help agents call Scribe correctly and interpret success, authentication, validation, network, and other failures. The structured output makes its commands easier to use in scripts.

Instructions file

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/naoray/scribe/claude-md
Clone the repo
git clone --depth 1 https://github.com/Naoray/scribe

Wrote 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.

agentmods badge for scribe CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/naoray/scribe/claude-md.svg)](https://agentmods.dev/instructions/naoray/scribe/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/naoray/scribe/claude-md"><img src="https://agentmods.dev/badge/instructions/naoray/scribe/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,743 This file is loaded in full into every session.
When invoked 1,743 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 $0.01743 $0.01743
Opus 5 $0.00872 $0.00872
Sonnet 5 $0.00349 $0.00349
Haiku 4.5 $0.00174 $0.00174

Measured 3d ago against content hash 9525e9dca409, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scribe 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 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.

CLAUDE.md · 137 lines

How it starts

The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Scribe Agent Contract

Scribe is an agent-first CLI for managing local AI coding agent skills. Prefer JSON mode for automation and inspect command schemas when composing calls.

JSON Envelope

Migrated commands emit a versioned envelope on stdout:

{
  "status": "ok",
  "format_version": "1",
  "data": {},
  "meta": {
    "duration_ms": 12,
    "bootstrap_ms": 3,
    "command": "scribe list",
    "scribe_version": "dev"
  }
}

Errors use the same envelope shape on stderr. Payload keys that were previously top-level now live under data.

Exit Codes

Code Meaning
0 success
1 general operational failure
2 usage or invalid flags
3 requested command, registry, skill, or schema was not found
4 permission or authentication failure
5 conflict requiring user action
6 network or remote service failure
7 temporarily unavailable local dependency
8 validation failure
9 user canceled
10 partial success; inspect data.summary.failed or item-level errors

Prompting Tips

Use scribe schema <command> --json before calling an unfamiliar command. Use --fields only on commands whose schema lists the flag. When updating scripts from pre-envelope output, change jq '.foo' to jq '.data.foo'.

Examples:

scribe list --json --fields name,status
scribe schema sync --json
scribe sync --json | jq '.data.summary'
scribe adopt --dry-run --json | jq '.data.conflicts'

Commands

Command Flags Output schema
scribe add --alias, --force, --id, --json, --no-interaction, --path, --ref, --registry, --repo, --resync, --source, --url yes
scribe adopt --dry-run, --json, --no-interaction, --verbose yes
scribe browse --id, --install, --json, --no-interaction, --path, --query, --ref, --registry, --repo, --resync, --source, --url no
scribe check --json yes
scribe config adoption --add-path, --json, --mode, --remove-path no
scribe config set editor --json no
scribe config set --json no
scribe config --json no
scribe create registry --json, --owner, --private, --repo, --team no
scribe create --json no
scribe doctor --fix, --json, --skill yes
scribe explain --json, --raw yes
scribe guide --json yes
scribe init --force, --json yes
scribe install --alias, --all, --force, --json, --registry no
scribe kit create --description, --force, --json, --mcp-servers, --registry, --skills no
scribe kit install --alias, --force, --json, --no-deps, --no-interaction no
scribe kit list --fields, --json, --local, --registry, --remote no
scribe kit push --json, --registry no
scribe kit show --json no
scribe kit sync --force, --json, --no-deps, --no-interaction no
scribe kit --json no
scribe list --fields, --json, --registry, --remote yes
scribe mcp list --json yes
scribe mcp --json no
scribe migrate global-to-projects --dry-run, --force, --json, --no-interaction, --project, --undo yes
scribe migrate --json no
scribe project init --force, --json, --kits yes
scribe project kit add --json, --no-sync yes
scribe project kit remove --json, --no-sync yes
scribe project kit --json no
scribe project snippet add --json, --no-sync yes
scribe project snippet remove --json, --no-sync yes
scribe project snippet --json no
scribe project skill add --json, --no-sync yes
scribe project skill claim --json yes
scribe project skill create --json yes
scribe project skill remove --block, --json, --no-sync yes
scribe project skill --json no
scribe project sync --check, --force, --json, --vendor yes
scribe project --json no
scribe push --json yes
scribe registry add --install, --json, --no-interaction, --registry no
scribe registry connect --force-kits, --id, --install-all, --json, --path, --ref, --repo, --source, --url yes
scribe registry create --json, --owner, --private, --repo, --team no
scribe registry disable --json no
scribe registry enable --json no
scribe registry forget --json no
scribe registry index --json no
scribe registry list --json no
scribe registry migrate --json no
scribe registry resync --force-kits, --json, --refresh-kits yes
scribe registry --json no
scribe remove --json, --no-interaction no
scribe resolve --json, --ours, --theirs no
scribe restore --json no
scribe schema --all, --json, --markdown no
scribe show --json no
scribe skill edit --add, --inherit, --json, --pin, --remove, --tools no
scribe skill repair --from, --json, --tool no
scribe skill tools --disable, --enable, --json, --reset no
scribe skill --json no
scribe status --json yes
scribe sync --alias, --force, --json, --registry, --trust-all yes
scribe tools add --detect, --install, --json, --path, --uninstall no
scribe tools disable --json no
scribe tools enable --json no
scribe tools --json no
scribe update --apply, --json yes
scribe upgrade-agent --json no
scribe upgrade --check, --json no
scribe no

Read the full file on GitHub · 137 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 · 137 lines · 1,743 tokens per session scan A 9525e9dca409

Subscribe to this mod's changes

scribe CLAUDE.md is an instructions file published in the GitHub repository Naoray/scribe (5 stars, last pushed 2mo ago), licensed MIT. It adds 1,743 tokens to every session, about $0.0087 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.