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/cvandesande/project-code-intelligence/agents-mdgit clone --depth 1 https://github.com/cvandesande/project-code-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 | $0.01419 | $0.01419 |
| Opus 5 | $0.00709 | $0.00709 |
| Sonnet 5 | $0.00284 | $0.00284 |
| Haiku 4.5 | $0.00142 | $0.00142 |
Grade A, and why
project-code-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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions
This repo has a project-code-intelligence MCP server. For non-trivial code discovery, use it before broad rg/find or speculative file reads. Use rg/direct reads for known small files and final verification.
Before you remove or rename a definition (function, class, method), or change
one's signature, call the blast_radius MCP tool for that symbol first. All
three break callers the same way, and the tool answers all three with one
query. Treat its callers, test coverage, and entry-point/orphan flags as
evidence (not a verdict), and verify live callers in source before finalizing
the change. The evidence hook injects the same bundle on an edit-delete as a
backstop, but pulling it up front is cheaper than undoing a bad removal.
To find duplication in the code already indexed, find_redundancy reports
groups of functions that repeat one call-shape, ranked by whether collapsing
them is worth it. For a whole-tree sweep (staleness, duplicate
names, redundancy candidates, static findings), run pci audit from
the repo root; measured precisions are printed in the report, and
docs/verified-duplications-2026-08-11.md carries the source-verified seed
list.
This repository builds project-code-intelligence: an MCP server that gives
coding agents structured access to indexed Git repositories — semantic
search, lexical search, a candidate-edge graph, and static-analysis findings
— so agents can find context without speculative file reads. Indexed data
lives in Postgres/pgvector.
Project Intent
- Keep the default package generic. Project-specific behaviour belongs in code profiles, not in shared ingest or MCP code.
- Do not publish or register private project profiles.
- Preserve the stdio MCP deployment path. Docker Compose runs local pgvector and optional embedding backends; it is not a deployment vehicle for the MCP server itself.
- Give a confidence level when recommending an approach, and push back when a suggestion would make the project less generic, less publishable, or unnecessarily complex.
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 · 123 lines · 1,419 tokens per session scan A 030a42f22059
project-code-intelligence AGENTS.md is an instructions file published in the GitHub repository cvandesande/project-code-intelligence (1 stars, last pushed 16d ago), licensed MIT. It adds 1,419 tokens to every session, about $0.0071 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
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.