contextception CLAUDE.md

Project instructions for Contextception, a Go tool that analyzes a codebase to determine which code an agent should understand before making a change. It is available as a command-line program and an MCP server, a service that exposes tools to AI agents.

In plain words
What is it for?
Use it to index and analyze Python, TypeScript or JavaScript, Go, Java, Rust, and C# repositories and produce context bundles for safe code changes.
Why use it?
It helps agents focus on relevant code and reduce the risk of changing a system without understanding its dependencies.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/kehoej/contextception/claude-md
Clone the repo
git clone --depth 1 https://github.com/kehoej/contextception
Per session 1,007 This file is loaded in full into every session.
When invoked 1,007 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.01007 $0.01007
Opus 5 $0.00504 $0.00504
Sonnet 5 $0.00201 $0.00201
Haiku 4.5 $0.00101 $0.00101

Measured 2d ago against content hash f86606fa2271, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

contextception 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 2d 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.

CLAUDE.md · 93 lines

How it starts

The opening of the file, as written. The whole thing — 93 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

Contextception is a code context intelligence engine written in Go. It answers: "What code must be understood before making a safe change?" It is not a code generator, AI assistant, or IDE — it determines what matters, not what to do.

Supports 6 languages: Python, TypeScript/JavaScript, Go, Java, Rust, and C#. Available as a CLI (16 commands) and MCP server (9 tools).

Tech Stack

  • Language: Go (single binary, concurrency for indexing)
  • Storage: SQLite via modernc.org/sqlite (pure-Go, embedded)
  • CLI: spf13/cobra
  • TS/JS Extraction: smacker/go-tree-sitter (CGO) with regex fallback
  • MCP Server: modelcontextprotocol/go-sdk (stdio transport)
  • Config: gopkg.in/yaml.v3
  • Semver: golang.org/x/mod/semver
  • Release Signing: jedisct1/go-minisign (signature verification for self-update)

Architecture

See docs/ARCHITECTURE.md for the full architecture guide with diagrams.

Repository → Language Extractors → SQLite Index → Analysis Engine → Context Bundle (JSON)

Key components: Indexer (internal/indexer/), Language Extractors (internal/extractor/), Resolvers (internal/resolver/), Analysis Engine (internal/analyzer/), Change Analysis (internal/change/), MCP Server (internal/mcpserver/), Update (internal/update/).

Key Design Principles

  1. Deterministic first — outputs are explainable and reproducible; no model-dependent core logic
  2. Static truth before AI — signals from dependency graphs, repo structure, change patterns
  3. Read-only by design — never modifies source, never executes arbitrary commands
  4. Incremental — only changed files reprocessed; schema migrations are versioned
  5. Pluggable extractors — language-specific extractors emit facts; core is language-agnostic

Development

make build      # Build binary to ./bin/contextception
make test       # Run all tests
make lint       # Run golangci-lint
make check      # Run vet + lint + test
make coverage   # Generate HTML coverage report
make install    # go install to $GOPATH/bin
make generate   # Regenerate protocol JSON schemas
make release    # Show release info (latest tag, next version, pending commits)
make help       # List all targets
/release        # Full release: generate changelog, commit, tag, push (Claude Code)

Read the full file on GitHub · 93 lines

Changes

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.

  1. 2d ago First seen · 93 lines · 1,007 tokens per session scan A f86606fa2271

Subscribe to this mod's changes

contextception CLAUDE.md is an instructions file published in the GitHub repository kehoej/contextception (6 stars, last pushed 1mo ago), licensed MIT. It adds 1,007 tokens to every session, about $0.0050 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.