AutoRAG Agent is a librarian-style system that searches files on a computer, checks the original documents, and organizes findings into numbered knowledge units using multiple retrieval methods. It helps an agent answer questions from local documents without first copying them into a centralized vector database, while the repository also maintains the earlier AutoRAG tool for optimizing retrieval-augmented-generation pipelines. The catalogue entries provide skills and instructions for working with AutoRAG Agent.
Borrowing it
Nothing to install: this file belongs to Marker-Inc-Korea/AutoRAG. 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/Marker-Inc-Korea/AutoRAG/main/AGENTS.mdgit clone --depth 1 https://github.com/Marker-Inc-Korea/AutoRAGWrote 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/marker-inc-korea/autorag/agents-md)<a href="https://agentmods.dev/instructions/marker-inc-korea/autorag/agents-md"><img src="https://agentmods.dev/badge/instructions/marker-inc-korea/autorag/agents-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/marker-inc-korea/autorag/agents-md"><img src="https://agentmods.dev/badge/instructions/marker-inc-korea/autorag/agents-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 | $0.13715 | $0.13715 |
| Opus 5.5 | $0.05486 | $0.05486 |
| Sonnet 5.5 | $0.02743 | $0.02743 |
| Haiku 4.5 | $0.01371 | $0.01371 |
Grade C, and why
AutoRAG AGENTS.md scanned grade C with 1 finding 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 today.
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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
`.autorag-e2e`; for manual cleanup, run `rm -rf .autorag-e2e` from this clone How it starts
The opening of the file, as written. The whole thing — 654 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AutoRAG — Pi-Powered Personal Librarian Agent
Guardrails
These bind every agent and human working in this checkout. Force is CI, CODEOWNERS, and the main rulesets; this file is the instruction. Do not flatten the rest of this map into these four boxes.
Allowed
Edits inside src/, test/, scripts/, skills/, internal-skills/, and docs/ that stay within the task.
Forbidden
- Changing the
LICENSElicense identifier - Adding or committing secrets (
.env,*.key, tokens, cookies, private corpus dumps) - Rewriting git history
- Attaching
node_modulesto releases - Sending corpus text to a remote provider without an explicit trusted-config opt-in (remote embedders and rerankers are supported; the local MinSync embedder remains the zero-config default)
- Hard-coding provider credentials in config files or argv; store only environment-variable, keychain, or profile references
- Creating git worktrees (this clone is the isolation boundary)
Required
bun run check && bun run lint && bun run typecheck, plus the tests that cover the change. Dependency or license-allowlist edits also need the supply-chain gate (bun run supply-chain).
Ask first
A new runtime dependency, a license-allowlist change, SECURITY / GOVERNANCE / AI policy, an embedding backend change or its license, and P2P default changes.
Git Workflow (binding)
This checkout is one of several clones of the same repository. Those clones are the isolation boundary. Never create a git worktree. Do not run git worktree add, do not create a linked checkout, and do not isolate a PR, review, or feature in a new worktree.
This section is the base workflow for this clone. If another agent skill, default, or habit (including worktree-based isolation) conflicts with it, follow this section.
Before any PR review, PR work, or new feature
git fetch origin.- Update local
mainto matchorigin/main(git checkout mainthengit pull --ff-only origin main). - Check out the branch that belongs to the work, in this clone:
- Existing PR or existing feature branch:
git checkout <branch>(orgh pr checkout <n>), then sync that branch with its remote upstream (git pull --ff-only). - New feature: create the branch from the just-updated
main(git checkout -b <branch>).
- Existing PR or existing feature branch:
- Only then edit, review, or test.
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.
- today Changed · +5 lines · +133 tokens per session e53a155b009d
- yesterday Changed · +20 tokens per session b9ab607d3fe9
- 2d ago Changed · +7 lines · +330 tokens per session 4b38814b14e2
- 3d ago Changed · +1 lines · +69 tokens per session 47d5555122e1
- 4d ago Changed · +36 lines · +752 tokens per session ae928ba317f7
- 5d ago Changed · -2 lines · -98 tokens per session d0f66e4a42ac
- 7d ago Changed · +3 lines · +433 tokens per session ef30bd538f57
- 9d ago Changed · +4 lines · +222 tokens per session cd5416550351
- 14d ago Changed · -2 lines · +59 tokens per session c4aaa99c1dd2
- 15d ago First seen · 602 lines · 11,795 tokens per session scan C a0bdc91842c6
AutoRAG AGENTS.md is an instructions file published in the GitHub repository Marker-Inc-Korea/AutoRAG (5,191 stars, last pushed yesterday), licensed Apache-2.0. It adds 13,715 tokens to every session, about $0.0549 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
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
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, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
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.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.