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/getledric/ledric/claude-mdgit clone --depth 1 https://github.com/getledric/ledricWrote 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/getledric/ledric/claude-md)<a href="https://agentmods.dev/instructions/getledric/ledric/claude-md"><img src="https://agentmods.dev/badge/instructions/getledric/ledric/claude-md.svg" alt="Measured on agentmods" 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 | $0.04235 | $0.04235 |
| Opus 5 | $0.02117 | $0.02117 |
| Sonnet 5 | $0.00847 | $0.00847 |
| Haiku 4.5 | $0.00424 | $0.00424 |
Grade A, and why
ledric 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.
How it starts
The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Repo shape
pnpm workspace, Node 22+, ESM-only TypeScript. Nine packages under packages/, plus a top-level e2e/ Playwright suite. All inter-package deps use workspace:* (rewritten to real versions at publish time). Native deps better-sqlite3 / sharp / esbuild are allow-listed for install scripts in pnpm-workspace.yaml — re-add them there if you ever pull native code in.
| Package | What it owns |
|---|---|
cli (published as ledric) |
The single binary. init, serve, http, get, ls, asset, keys, tag, types, … |
core |
The brain — read/find/draft/publish/alterType/describeModel. No I/O directly; takes a Storage plus an optional TransformCache, httpBase, auth. |
schema |
Pure (no I/O): defineType(), field.* builders, FIELD_TYPE_SPECS (the field-type catalogue including the wire_shape advertisement), validator. Imported by both consumer code and core itself. |
storage |
Dialect-and-adapter pattern: same logical operations (createType, findEntries, searchEntries, addEntryTags, …) implemented for SQLite (default) / Postgres / MySQL via Kysely. |
mcp-server |
The 20-tool MCP surface plus SERVER_INSTRUCTIONS. Tool args are zod schemas; JSON Schemas are written separately for the ListTools response. |
http-server |
Fastify routes that mirror the MCP catalogue at POST /rpc, plus dedicated REST endpoints for reads, asset bytes, and asset uploads. Auth is per-tool, not per-method (READ_RPC_TOOLS set decides which tools accept reader keys via /rpc). |
gui |
React+htm SPA. Static files only — web/* is shipped as-is and rendered by the browser via CDN imports + htm/react (no JSX, no bundler). Only the package's src/index.ts is tsup-built. |
sdk |
TypeScript consumer client. LedricEntries is an empty interface that ledric types --augment-sdk augments. |
proxy |
Server-side (Request) => Promise<Response> primitive for consumer sites. Strips inbound auth headers and injects the configured Bearer; gates writes; framework-agnostic. |
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.
- 3d ago First seen · 144 lines · 4,235 tokens per session scan A 6035d740e566
ledric CLAUDE.md is an instructions file published in the GitHub repository getledric/ledric (0 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 4,235 tokens to every session, about $0.0212 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.
Other instructions, from other repositories
basic-blog AGENTS.md
Instructions for daocodotorg/basic-blog, covering agent notes — basic-blog monorepo, scope, npm release (automatic when version changes), when to tag and commands (agent should run these unless the user opts out).
sonicjs AGENTS.md
AGENTS.md instructions for SonicJs-Org/sonicjs, covering sonicjs agent guidelines, decision tree — code exploration, rtk (rust token killer), caveman mode and delegation — cavecrew subagents.
sonicjs CLAUDE.md
Claude Code instructions for SonicJs-Org/sonicjs, covering sonicjs ai development guidelines, core technology stack, workspace boundary (conductor), architecture direction: document model (authoritative) and the 5 document tables (migration 0002documents.sql).
flowwink CLAUDE.md
Claude Code instructions for magnusfroste/flowwink, covering claude.md, commands, development (auto-runs migrations before starting), production build and lint.
vibecheck CLAUDE.md
Claude Code instructions for EpleCheck/vibecheck, covering vibecheck — repo guide, requirements, layout (npm-workspaces monorepo), setup & commands and editing content (the main task).
NavoCMS AGENTS.md
AGENTS.md instructions for Tsumibito/NavoCMS, covering navocms agent contract, product invariants, contract changes and working practices.