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/denfry/codebase-index/agents-mdgit clone --depth 1 https://github.com/denfry/codebase-indexWhat 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 | $0.00614 | $0.00614 |
| Opus 5 | $0.00307 | $0.00307 |
| Sonnet 5 | $0.00123 | $0.00123 |
| Haiku 4.5 | $0.00061 | $0.00061 |
Grade A, and why
codebase-index 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
codebase-index
Use the local codebase index before scanning repository files.
Skill resources: .codex/skills/codebase-index/SKILL.md
Run codebase-index search "<query>" --json for general questions, or use
symbol, refs, impact, and graph for symbol lookup, references, change
impact, and HTML graph export. Search/read commands auto-build the index when
it is missing; run codebase-index update when responses report stale data.
Versioning Policy
Use Semantic Versioning in the MAJOR.MINOR.PATCH format.
- For fixes, maintenance, documentation, tests, small features, and other
changes that are not broad or substantially alter the product, increment
only
PATCH:1.2.3->1.2.4. - For broad or substantial changes, new major capabilities, or notable changes
to behavior or architecture, increment
MINORand resetPATCH:1.2.3->1.3.0. - Do not increment
MAJOR(for example,1.x.x->2.0.0) unless the user explicitly requests it or explicitly approves a breaking release. - When uncertain, use a
PATCHincrement and do not overstate the scope of the release. - When changing the version, update every canonical version location and the changelog together so release metadata remains consistent.
Fork and Pull Request Workflow
When contributing from a fork, keep the pull request focused and easy to review.
- Add the canonical repository as
upstream; keep the contributor fork asorigin. - Never commit directly to
main. Create a branch from the latestupstream/mainusing<type>/<short-description>. - Keep unrelated changes in separate branches and pull requests. Do not include generated files, build artifacts, local configuration, credentials, or editor-specific files.
- Rebase the branch onto the latest
upstream/mainbefore opening or updating the pull request. Do not mergemaininto the feature branch unless a maintainer explicitly requests it. - Run the relevant tests, linting, formatting, and type checks before pushing.
- Use Conventional Commits and write a pull request description that explains the problem, the solution, verification performed, and any compatibility or migration impact.
- Add user-visible changes to
CHANGELOG.mdunder[Unreleased], but do not change the project version unless a maintainer explicitly requests a release version bump. - Push only the contributor branch to the fork, then open a pull request against
the canonical repository's
mainbranch. Never force-push after review has started unless rewriting the branch is necessary and clearly communicated.
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 · 58 lines · 614 tokens per session scan A 3e68e12d44cc
codebase-index AGENTS.md is an instructions file published in the GitHub repository denfry/codebase-index (7 stars, last pushed 1mo ago), licensed MIT. It adds 614 tokens to every session, about $0.0031 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
leantoken AGENTS.md
Instructions for morluto/leantoken, covering repository guidance, architecture, development, change-specific validation and contributions.
claude-context CLAUDE.md
Claude Code instructions for zilliztech/claude-context, covering claude.md, monorepo layout, commands, tests and running the mcp server locally.
open-codebase-index AGENTS.md
Instructions for Helweg/open-codebase-index, covering agents.md - ai agent guidelines for open-codebase-index, build, test, and lint, run a single test, equivalent low-level command and architecture and file structure.
paparats-mcp CLAUDE.md
Claude Code instructions for IBazylchuk/paparats-mcp, covering paparats-mcp, ids, architecture, typescript conventions and module structure.
mimirs CLAUDE.md
Instructions for TheWinci/mimirs, a project described as: Local MCP server that gives AI coding agents persistent, searchable memory of your codebase.
ContextAtlas AGENTS.md
Instructions for codefromkarl/ContextAtlas, covering agents.md, project overview, development commands, build and development.