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/r-darwish/daruma/agents-mdgit clone --depth 1 https://github.com/r-darwish/darumaWrote 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/r-darwish/daruma/agents-md)<a href="https://agentmods.dev/instructions/r-darwish/daruma/agents-md"><img src="https://agentmods.dev/badge/instructions/r-darwish/daruma/agents-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 5.1 | $0.01092 | $0.01092 |
| Opus 5 | $0.00546 | $0.00546 |
| Sonnet 5 | $0.00218 | $0.00218 |
| Haiku 4.5 | $0.00109 | $0.00109 |
Grade A, and why
daruma AGENTS.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 — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guidelines for AI Coding Agents (AGENTS.md)
Welcome, agent. You are tasked with maintaining or extending Daruma (the contextual lesson-learned MCP). Please adhere to the following tactical guidelines to ensure code quality, system safety, and strict alignment across development cycles.
CRITICAL INSTRUCTION: Dynamic Documentation Updates
[!IMPORTANT] As features, CLI flags, database schemas, or resolver logic are updated or added, you MUST immediately update both README.md and AGENTS.md. Do not leave files lagging behind the actual implementation. State clearly in your PR or walkthrough what updates were made to these documents.
Technical Design Overview
Daruma resolves a developer's directory path into a multi-layered, prioritized coordinate:
- Component Layer (
(repo, component)): Folder containing descriptor files likego.modorcargo.tomlin the repository tree. - Repository Layer (
(repo, 'root')): Root of the repository containing.gitor.jj. - Technology Layer (
tech): Detected tech stack (e.g.go,rust,python,typescript,docker) based on markers. - Global Layer (
'global', 'global'): General, project-agnostic development patterns.
Storage Rules
- Database resides at
~/.local/share/daruma/daruma.db. - Database Concurrency: The SQLite connection must enable WAL (Write-Ahead Logging) mode and use a normal sync flag. This ensures readers do not block writers and vice-versa, allowing safe multi-process local execution.
- Pure Functions: Ensure database and resolver functions are designed as pure, side-effect-free, easily mockable functions.
Version Control Agnosticism
- Daruma supports standard Git and Jujutsu (
jj). - Check if
.jjis present in the ancestor tree. If so, invokejjCLI wrappers instead ofgitfor remote and log operations.
Coding Guidelines
- Test-Driven Design:
- Every core resolver or database capability must have dedicated
pytesttests. - Use
pytestparameterized tests (pytest.mark.parametrize) to enforce exhaustive boundary and type coverage.
- Every core resolver or database capability must have dedicated
- Type Hinting: All Python code must utilize modern type annotations (
typing.Tuple,typing.Optional,typing.List, etc.). - Dependency Management: Use
uv addto inject external dependencies. Keep dependencies strictly to the bare minimum (mcp,click,pytestis the default). - CLI Development: The CLI tool (
daruma) must wrap$EDITORsafely. It should write templates to temporary Markdown files and retrieve them only after the process exits successfully.
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 · 74 lines · 1,092 tokens per session scan A 358e8bd4be81
daruma AGENTS.md is an instructions file published in the GitHub repository r-darwish/daruma (0 stars, last pushed 27d ago), licensed MIT. It adds 1,092 tokens to every session, about $0.0055 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
aeon CLAUDE.md
Claude Code instructions for aeonfun/aeon, covering aeon, how aeon works, strategy, voice and soul file hierarchy (read in this order).
cognirepo CLAUDE.md
Claude Code instructions for ashlesh-t/cognirepo, covering claude.md, key rules, session start sequence (run in this order), behavioral confirmation rule and personas (cognirepo-402, cognirepo-403).
wayland-core copilot-instructions.md
Copilot instructions for FerroxLabs/wayland-core, covering ijfw rules, output discipline, memory routing, context discipline and cross-audit.
mcp-structured-memory CLAUDE.md
Claude Code instructions for nmeierpolys/mcp-structured-memory, a project described as: Structured Memory MCP Server.
inkwell-memory CLAUDE.md
Instructions for veronchenko/inkwell-memory, covering claude.md — inkwellmemory, layout, multi-tenant mode (inkwellmultitenant=1), conventions and testing.
memory-mcp-1file AGENTS.md
Instructions for pomazanbohdan/memory-mcp-1file, covering ⛔ l0 invariants (never violate under any circumstance), detection heuristic, 🔀 phase transition routing (⛔ blocking), gate-0: phase identification & loading and 🧾 proof-of-load requirement (critical).