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/clouatre-labs/aptu-coder/agents-mdgit clone --depth 1 https://github.com/clouatre-labs/aptu-coderWhat 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 | $0.01772 | $0.01772 |
| Opus 5 | $0.00886 | $0.00886 |
| Sonnet 5 | $0.00354 | $0.00354 |
| Haiku 4.5 | $0.00177 | $0.00177 |
Grade A, and why
aptu-coder 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 yesterday.
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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Project structure
Rust workspace with two crates:
crates/aptu-coder-core-- parsing, analysis, formatting, graph, pagination, typescrates/aptu-coder-- MCP server, tool handlers, logging, metrics, MCP Resources surface (list_resources, list_resource_templates, read_resource in crates/aptu-coder/src/tools/resources.rs)
MCP tool surface and parameter schemas are documented by the tools themselves; do not duplicate them here. Tool handler logic lives in crates/aptu-coder/src/tools/<tool>.rs.
Rust edition 2024, async with tokio, latest MCP protocol via rmcp. Supported languages are listed in crates/aptu-coder-core/src/lang.rs.
CI runners
All CI jobs run on ubuntu-24.04-arm (ARM64). Build, test, lint, and release jobs all target this image.
Commands
cargo build
cargo test
cargo clippy -- -D warnings
cargo fmt --check
cargo deny check advisories licenses
cargo bench
cargo install --path crates/aptu-coder --profile release # local install; binary lands in ~/.cargo/bin/
Workspace lints enforced in CI (deny): undocumented_unsafe_blocks, unwrap_used, expect_used. Both unwrap_used and expect_used have test-code exemptions via cfg_attr in each crate's lib.rs. Dependency freshness: new Cargo.lock entries must be >=7 days old; bypass with SKIP_PACKAGE_AGE_CHECK=true.
Observability
Two parallel telemetry channels; neither blocks tool execution.
- JSONL (always-on): daily-rotated files at
$XDG_DATA_HOME/aptu-coder/metrics-YYYY-MM-DD.jsonl; 30-day retention. See docs/METRICS.md for schema and span attribute policy. - OpenTelemetry (opt-in): set
OTEL_EXPORTER_OTLP_ENDPOINTto enable OTLP/HTTP export; noop providers when unset. W3C Trace Context extracted from MCP_metaso tool spans appear as children in the calling agent's distributed trace.
JSONL Metrics Analysis
Files: $XDG_DATA_HOME/aptu-coder/metrics-YYYY-MM-DD.jsonl (default: ~/.local/share/aptu-coder/). Full schema in docs/METRICS.md.
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.
- yesterday First seen · 89 lines · 1,772 tokens per session scan A 0eb9e2d74a2c
aptu-coder AGENTS.md is an instructions file published in the GitHub repository clouatre-labs/aptu-coder (5 stars, last pushed yesterday), licensed Apache-2.0. It adds 1,772 tokens to every session, about $0.0089 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
codeweave-mcp CLAUDE.md
Instructions for semihkayan/codeweave-mcp, covering claude.md, what this project is, build & run, architecture and layers.
codexray CLAUDE.md
Instructions for NeuralRays/codexray, covering codexray — ai agent instructions, workflow and key principle.
graphify AGENTS.md
AGENTS.md instructions for Graphify-Labs/graphify: This project has a graphify knowledge graph at graphify-out/.
codeseek CLAUDE.md
Instructions for CodeBendKit/codeseek, covering claude.md, build & test, enter rust project directory, build and run tests.
arbor CLAUDE.md
Claude Code instructions for getArbor-dev/arbor, covering claude.md, commands, build, test all crates and test single crate.
bonsai-ninja AGENTS.md
Instructions for gromhacks/bonsai-ninja, covering bonsai-ninja, map a codebase, optional explicit semantic sidecar prewarm, explicit spelling for default syntax/construct indexing and optional during active editing.