codespelunker, invoked as cs, is a code-search tool that parses repositories to distinguish executable code from comments and strings, then ranks matches by relevance. Developers and AI agents use it to locate implementations and useful snippets through a CLI, terminal interface, or HTTP service without maintaining a search index. The catalogue instruction helps coding agents use its structured search capabilities.
Borrowing it
Nothing to install: this file belongs to boyter/cs. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/boyter/cs/master/CLAUDE.mdgit clone --depth 1 https://github.com/boyter/csWrote 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/boyter/cs/claude-md)<a href="https://agentmods.dev/instructions/boyter/cs/claude-md"><img src="https://agentmods.dev/badge/instructions/boyter/cs/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/boyter/cs/claude-md"><img src="https://agentmods.dev/badge/instructions/boyter/cs/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.01022 | $0.01022 |
| Opus 5.5 | $0.00409 | $0.00409 |
| Sonnet 5.5 | $0.00204 | $0.00204 |
| Haiku 4.5 | $0.00102 | $0.00102 |
Grade A, and why
cs 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 2mo 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
cs (codespelunker) is a command-line code search tool written in Go. It searches files recursively using boolean queries, regex, and fuzzy matching, with relevance ranking (BM25/TF-IDF). Four modes: console output, interactive TUI, HTTP server, and MCP server.
Build & Test Commands
go build -o cs # Build binary
go test ./... # Run all tests
go test -v ./... # Verbose tests
go test -run TestPreParseQuery ./... # Run a single test
Linting:
golangci-lint run --enable=gofmt ./...
gofmt -s -w -l .
Architecture
Root package main plus subpackages under pkg/. The search pipeline is orchestrated by DoSearch() in search.go:
FileWalker → [read + filter files] → AST query evaluation → matched FileJobs channel
Query Pipeline (pkg/search/)
Queries are parsed into an AST and evaluated against each file:
Lexer → Parser → Transformer → Planner → Executor
- Lexer (
lexer.go): tokenizes query string (terms, quotes, regex, operators, fuzzy markers) - Parser (
parser.go): builds AST with boolean logic (AND/OR/NOT), quoted phrases, regex/pattern/, fuzzy~1/~2 - Transformer (
transformer.go): semantic rewrites (e.g.complexity=high) - Planner (
planner.go): query optimization - Executor (
executor.go): evaluates AST against file content, returns match locations - Extractor (
extractor.go): extracts matched terms for highlighting - AST (
ast.go): AST node type definitions (AndNode, OrNode, NotNode, KeywordNode, PhraseNode, RegexNode, FuzzyNode, FilterNode) - Document (
document.go): Document and SearchResult structs
See pkg/search/README.md for detailed query syntax documentation.
Core Root Files
main.go: Cobra CLI entry point. Routes toConsoleSearch(), TUI (initialModel()),StartHttpServer(), orStartMcpServer()based on flags/argsconfig.go:Configstruct with all CLI-configurable fields.DefaultConfig()provides sensible defaultssearch.go:DoSearch()— orchestrates the full search pipeline: file walking, reading, binary/minified filtering, AST evaluation, and result streaming via channels. UsesSearchCachefor prefix-based cachingconsole.go:ConsoleSearch()— collects results, ranks, and outputs in text/JSON/vimgrep formattui.go: bubbletea TUI with lipgloss styling. Debounced search input, incremental result streaming, syntax-highlighted previewhttp.go: HTTP server with embedded templates, search/display endpoints, theme support (dark/light/bare), custom template overridescache.go:SearchCache— LRU cache with TTL for query results; supports prefix matching for progressive refinement in TUIsyntax.go: Syntax highlighting with keyword tables for 80+ languageslanguage.go: Language detection via scc processor's language databasemcp.go: MCP (Model Context Protocol) server over stdio; exposes search as a tool for AI agentstemplates.go: Template loading with//go:embedfor built-in HTML templates; supports custom template paths
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.
- 2mo ago First seen · 78 lines · 1,022 tokens per session scan A a0d56afc0a09
cs CLAUDE.md is an instructions file published in the GitHub repository boyter/cs (1,080 stars, last pushed 10d ago), licensed MIT. It adds 1,022 tokens to every session, about $0.0041 per session on Opus 5.5. 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.
cs CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| cs CLAUDE.mdboyter/cs | Instructions file | 1.1k | 10d ago | MIT | A | 1,022 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |