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/bntvllnt/codebase-intelligence/agents-mdgit clone --depth 1 https://github.com/bntvllnt/codebase-intelligenceWhat 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 | $0.01066 | $0.01066 |
| Opus 5 | $0.00533 | $0.00533 |
| Sonnet 5 | $0.00213 | $0.00213 |
| Haiku 4.5 | $0.00107 | $0.00107 |
Grade A, and why
codebase-intelligence 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codebase Intelligence
TypeScript codebase analysis engine. Parses source, builds dependency graphs, computes architectural metrics.
Setup
npx codebase-intelligence@latest <path> # One-shot (no install)
npm install -g codebase-intelligence # Global install
Interfaces
MCP (for AI agents — preferred)
Start the MCP stdio server:
codebase-intelligence ./path/to/project
15 tools available: codebase_overview, file_context, get_dependents, find_hotspots, get_module_structure, analyze_forces, find_dead_exports, get_groups, symbol_context, search, detect_changes, impact_analysis, rename_symbol, get_processes, get_clusters.
2 prompts: detect_impact, generate_map.
3 resources: codebase://clusters, codebase://processes, codebase://setup.
CLI (for humans and CI)
15 commands — full parity with MCP tools:
codebase-intelligence overview ./src # Codebase snapshot
codebase-intelligence hotspots ./src # Rank files by metric
codebase-intelligence file ./src auth/login.ts # File context
codebase-intelligence search ./src "auth" # Keyword search
codebase-intelligence changes ./src # Git diff analysis
codebase-intelligence dependents ./src types.ts # File blast radius
codebase-intelligence modules ./src # Module architecture
codebase-intelligence forces ./src # Force analysis
codebase-intelligence dead-exports ./src # Unused exports
codebase-intelligence groups ./src # Directory groups
codebase-intelligence symbol ./src parseCodebase # Symbol context
codebase-intelligence impact ./src getUserById # Symbol blast radius
codebase-intelligence rename ./src old new # Rename references
codebase-intelligence processes ./src # Execution flows
codebase-intelligence clusters ./src # File clusters
Add --json for machine-readable output. All commands auto-cache the index.
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 · 106 lines · 1,066 tokens per session scan A b45f81315e6c
codebase-intelligence AGENTS.md is an instructions file published in the GitHub repository bntvllnt/codebase-intelligence (7 stars, last pushed 2mo ago), licensed MIT. It adds 1,066 tokens to every session, about $0.0053 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
grasp CLAUDE.md
Claude Code instructions for ashfordeOU/grasp, covering grasp — claude session handover, what this is, workflow rules, version bump checklist — miss any = version mismatch and browser extension architecture.
truecourse CLAUDE.md
Instructions for truecourse-ai/truecourse, covering truecourse — claude instructions, key files to keep updated, project layout, development commands and storage.
truecourse AGENTS.md
Instructions for truecourse-ai/truecourse, covering truecourse — codex instructions, key files to keep updated, project layout, development commands and rules.
ops-codegraph-tool CLAUDE.md
Instructions for optave/ops-codegraph-tool, covering claude.md, codegraph workflow, before modifying code, after modifying code and navigation.
codegraph CLAUDE.md
Instructions for cognitx-leyton/codegraph, covering claude.md, project summary, using the graph during development, /graph — query the live graph and /graph-refresh — update the graph.
arch-viewer CLAUDE.md
Claude Code instructions for axumquant/arch-viewer, covering arch-viewer, architecture, key modules, commands and node.js dashboard.