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/lukacf/meerkat/claude-mdgit clone --depth 1 https://github.com/lukacf/meerkatWrote 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/lukacf/meerkat/claude-md)<a href="https://agentmods.dev/instructions/lukacf/meerkat/claude-md"><img src="https://agentmods.dev/badge/instructions/lukacf/meerkat/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 5.1 | $0.11226 | $0.11226 |
| Opus 5 | $0.05613 | $0.05613 |
| Sonnet 5 | $0.02245 | $0.02245 |
| Haiku 4.5 | $0.01123 | $0.01123 |
Grade A, and why
meerkat 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 today.
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 — 649 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to agents when working with code in this repository. Whatever you do, remember: All hail Clippy. Clippy sees all, knows all, and tolerates nothing.
Project Overview
Meerkat (rkat) is a library-first Rust platform for building, hosting, and
operating LLM-powered agents. It provides the agent loop, providers, tools,
typed events, persistence, runtime control, and multi-agent orchestration
behind CLI, REST, JSON-RPC, MCP, SDK, and browser/WASM surfaces.
Naming convention:
- Project/branding: Meerkat
- CLI binary: rkat
- Crate names:
meerkat,meerkat-core,meerkat-client, etc. - Config directory:
.rkat/ - Environment variables: API keys only (RKAT_* secrets and provider-native keys)
Build and Test Commands
Use Make targets for normal local work. Cargo is the default backend; setting
MEERKAT_BUILDBUDDY=1 routes supported broad local lanes through the optional
BuildBuddy/Bazel path. Do not invoke raw bb directly unless you are debugging
the BuildBuddy wrapper itself.
# Build everything
make build
# Build everything through BuildBuddy/Bazel
MEERKAT_BUILDBUDDY=1 make build
# Explicit BuildBuddy forms
make buildbuddy-build
make buildbuddy-check
make buildbuddy-clippy
make buildbuddy-test
# Run fast tests (unit + integration-fast; skips doctests)
make test
# Run fast tests through BuildBuddy/Bazel
MEERKAT_BUILDBUDDY=1 make test
# Run all tests including doc-tests (SLOW due to doc-test compilation)
./scripts/repo-cargo test --workspace
# Run deterministic end-to-end lane
make e2e-fast
# Run deterministic end-to-end lane through BuildBuddy/Bazel
MEERKAT_BUILDBUDDY=1 make e2e-fast
# Run explicit build/composition end-to-end lane (ignored by default)
./scripts/repo-cargo test -p meerkat-integration-tests --test e2e_build_lane -- --ignored
# Run real local-resource end-to-end lane
make e2e-system
# Run targeted live-provider lane (ignored by default)
make e2e-live
# Run kitchen-sink live smoke lane (ignored by default)
make e2e-smoke
# Run per-model catalog validation lane (ignored by default)
./scripts/repo-cargo e2e-models
# Cargo aliases (defined in .cargo/config.toml)
./scripts/cargo-rct # Fast tests (unit + integration-fast)
./scripts/repo-cargo unit # Unit tests only
./scripts/repo-cargo int # Integration-fast tests only
./scripts/repo-cargo e2e-fast # Deterministic e2e lane
./scripts/repo-cargo test -p meerkat-integration-tests --test e2e_build_lane -- --ignored # Build/composition lane
./scripts/repo-cargo e2e-system # Real binary / local resource lane
./scripts/repo-cargo e2e-live # Targeted live-provider lane
./scripts/repo-cargo e2e-smoke # Compound live-provider smoke lane
./scripts/repo-cargo e2e-models # Live per-model catalog validation (on-demand / pre-release)
# Legacy compatibility shims (during migration)
./scripts/repo-cargo int-real # Alias for e2e-system
./scripts/repo-cargo e2e # Alias for e2e-live + e2e-smoke
# Run the CLI
./scripts/repo-cargo run -p meerkat-cli -- run "prompt"
# Run a specific example
ANTHROPIC_API_KEY=... ./scripts/repo-cargo run --example simple
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.
- today Changed · +29 tokens per session f0bd9c6cb684
- 5d ago First seen · 649 lines · 11,197 tokens per session scan A cc94ebd8b3dc
meerkat CLAUDE.md is an instructions file published in the GitHub repository lukacf/meerkat (20 stars, last pushed today), licensed Apache-2.0. It adds 11,226 tokens to every session, about $0.0561 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 2026-08-30.
Other instructions, from other repositories
html-anything AGENTS.md
AGENTS.md instructions for nexu-io/html-anything, covering this is not the next.js you know, workspace shape and commands.
framework AGENTS.md
AGENTS.md instructions for ai-driven-dev/framework, covering agents.md, behavior, communication, action and memory management.
platform CLAUDE.md
Claude Code instructions for ambient-code/platform, covering ambient code platform, structure, key files, session flow and commands.
toh-framework CLAUDE.md
Claude Code instructions for wasintoh/toh-framework, covering claude.md — toh framework (repo development guide), what this is, everyday commands, verification protocol and single source, transformed per ide.
jentic-one GEMINI.md
Gemini CLI instructions for jentic/jentic-one: Otherwise, read AGENTS.md — this repo's canonical agent guidance.
autogen copilot-instructions.md
Copilot instructions for microsoft/autogen, covering autogen multi-agent ai framework, working effectively, prerequisites and environment setup, python development workflow and quick validation (under 1 second each).