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/vedantggwp/grimoire/claude-mdgit clone --depth 1 https://github.com/vedantggwp/grimoireWrote 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/vedantggwp/grimoire/claude-md)<a href="https://agentmods.dev/instructions/vedantggwp/grimoire/claude-md"><img src="https://agentmods.dev/badge/instructions/vedantggwp/grimoire/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 | $0.00826 | $0.00826 |
| Opus 5 | $0.00413 | $0.00413 |
| Sonnet 5 | $0.00165 | $0.00165 |
| Haiku 4.5 | $0.00083 | $0.00083 |
Grade A, and why
grimoire 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 4d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
You are in Grimoire, a local-first knowledge base builder packaged as a Claude Code plugin. Grimoire turns a topic into a structured knowledge base by scouting sources, ingesting and compiling them into wiki articles, generating a study-oriented frontend, and optionally serving the result to LLM tools through MCP.
This file loads on every conversation. Keep it lean.
- Read
SOUL.mdfor the full product bible, architecture, boundaries, or source-of-truth clarification - Read
docs/decisions.mdfor architectural decision history
Plugin Structure
Grimoire is a Claude Code plugin. Skills are auto-discovered from skills/*/SKILL.md.
| Skill | Job |
|---|---|
run |
One-command pipeline: new → scout → ingest → compile → present, 2 checkpoints |
new |
Interactive questionnaire + workspace scaffolding |
scout |
Research sources, score confidence, curate URL list |
ingest |
Fetch sources, preserve raw text, compile wiki articles |
compile |
Cross-references, backlinks, overview, gap analysis |
present |
Generate study-oriented static frontend |
serve |
Expose knowledge via custom MCP server |
update |
Headless scheduled refresh: delta scout → policy ingest → connections → freshness → PR |
Core Rules
-
One skill, one job. Keep responsibilities isolated. Each skill has its own references and assets.
-
Plain text is the interface. Markdown files are the contract between skills. Prefer transparent, editable artifacts.
-
Keep outputs editable and local. Humans must be able to inspect and modify outputs between steps.
-
Use
${CLAUDE_PLUGIN_ROOT}for all internal path references. Never hardcode absolute paths. -
Papyr Core is a plugin dependency. It handles markdown parsing, graph construction, search indexing, and analytics. Do not rebuild what it provides.
Workspace Convention
Maintain MANIFEST.md as the workspace change ledger. Update it whenever files are created, edited, renamed, or deleted.
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.
- 4d ago First seen · 75 lines · 826 tokens per session scan A 1a240659dbad
grimoire CLAUDE.md is an instructions file published in the GitHub repository vedantggwp/grimoire (2 stars, last pushed 1mo ago), licensed MIT. It adds 826 tokens to every session, about $0.0041 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
Vault-Agent-Memory AGENTS.md
Instructions for zycaskevin/Vault-Agent-Memory, covering agent instructions for vault agent memory, common install architecture, second decision: optional features, first decision: database scope and safe agent workflow.
memobase AGENTS.md
Instructions for maximskorohod/memobase, covering agents.md — установка плагина memobase, что это, способ b — через git (человек ставит вручную, без чата), безопасность and справочные факты (для сверки при установке).
llm-wiki-agent GEMINI.md
Gemini CLI instructions for SamurAIGPT/llm-wiki-agent, covering llm wiki agent — schema & workflow instructions, how to use, directory layout, page format and ingest workflow.
gini-agent AGENTS.md
AGENTS.md instructions for Open-Curiosity/gini-agent, covering gini agent instructions, shape, adrs, boundaries and branches.
remnic AGENTS.md
AGENTS.md instructions for joshuaswarren/remnic, covering remnic - agent guide, architecture boundaries (non-negotiable), upstream references, adapter implementation rules and openclaw compatibility window.
plur CLAUDE.md
Claude Code instructions for plur-ai/plur, covering claude.md, what is plur, development, package dependency and version bumps.