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/bahdotsh/indxr/claude-mdgit clone --depth 1 https://github.com/bahdotsh/indxrWrote 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/bahdotsh/indxr/claude-md)<a href="https://agentmods.dev/instructions/bahdotsh/indxr/claude-md"><img src="https://agentmods.dev/badge/instructions/bahdotsh/indxr/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 | $0.03890 | $0.03890 |
| Opus 5 | $0.01945 | $0.01945 |
| Sonnet 5 | $0.00778 | $0.00778 |
| Haiku 4.5 | $0.00389 | $0.00389 |
Grade A, and why
indxr 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 6d 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 — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.
indxr
A fast Rust codebase indexer for AI agents. Extracts structural maps (declarations, imports, tree) using tree-sitter and regex parsing across 27 languages.
Codebase Navigation — MUST USE indxr MCP tools
An MCP server called indxr is available. Always use indxr tools before the Read tool. Do NOT read full source files as a first step — use the MCP tools to explore, then read only what you need.
Token savings reference
The MCP server defaults to 3 compound tools (find, summarize, read). All 26 tools (3 compound + 23 granular) are available with --all-tools. With --features wiki, 9 additional wiki tools are available.
| Action | Approx tokens | When to use |
|---|---|---|
find(query) |
~100-400 | Find files/symbols by concept, name, callers, or signature pattern |
summarize(path) |
~200-600 | Understand a file, batch of files, or symbol without reading source |
read(path, symbol?) |
~50-300 | Read one function/struct. Supports symbols array and collapse. |
Read (full file) |
500-10000+ | ONLY when editing or need exact formatting |
Typical exploration: ~500 tokens vs ~3000+ for reading a full file (6x reduction).
Exploration workflow (follow this order)
The default 3 compound tools cover the most common exploration patterns:
find(query)— find files/symbols by concept, partial name, or type pattern. Start here when you know what you're looking for but not where it is.- Default mode (
relevant): multi-signal relevance search across paths, names, signatures, and docs. Supportskindfilter. mode: "symbol": find declarations by name (case-insensitive substring).mode: "callers": find who references a symbol (imports + signatures).mode: "signature": find functions by signature pattern (e.g.,"-> Result<").
- Default mode (
summarize(path)— understand files and symbols without reading source code.- File path (e.g.,
"src/main.rs"): complete file overview (declarations, imports, counts). - Glob pattern (e.g.,
"src/mcp/*.rs"): batch summaries for multiple files. - Symbol name (no
/, e.g.,"Cache"): full interface details (signature, doc comment, relationships). scope: "public": show only public API surface.
- File path (e.g.,
read(path, symbol?)— read source code by symbol name or explicit line range. Cap: 200 lines. Usesymbolsarray to read multiple in one call (500 line cap). Usecollapse: trueto fold nested bodies.
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.
- 6d ago First seen · 255 lines · 3,890 tokens per session scan A a5e49952e9e2
indxr CLAUDE.md is an instructions file published in the GitHub repository bahdotsh/indxr (72 stars, last pushed 5mo ago), licensed MIT. It adds 3,890 tokens to every session, about $0.0194 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
Archive-Agent CLAUDE.md
Claude Code instructions for shredEngineer/Archive-Agent, covering claude code development guide for archive agent, key architecture components, core processing pipeline, important modules and testing and quality assurance.
Trello-Desktop-MCP CLAUDE.md
Instructions for kocakli/Trello-Desktop-MCP, covering claude.md, project overview, build & development commands, architecture and entry point & protocol.
CRUX AGENTS.md
AGENTS.md instructions for JustineDevs/CRUX, covering crux, project overview, tech stack, frontend and backend.
CRUX CLAUDE.md
Claude Code instructions for JustineDevs/CRUX, covering crux, project overview, tech stack, frontend and backend.
prompt-control-plane CLAUDE.md
Claude Code instructions for rishi-banerjee1/prompt-control-plane, covering pcp -- prompt control plane, what this is, architecture, core pipeline (8 phases) and entry points.
audio-transcriber AGENTS.md
Instructions for Knuckles-Team/audio-transcriber, covering agents.md, tech stack & architecture, architecture diagram, workflow diagram and commands (run these exactly).