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/bartoszarendt/agenticloop/agents-mdgit clone --depth 1 https://github.com/bartoszarendt/agenticloopWhat 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 | $0.01980 | $0.01980 |
| Opus 5 | $0.00990 | $0.00990 |
| Sonnet 5 | $0.00396 | $0.00396 |
| Haiku 4.5 | $0.00198 | $0.00198 |
Grade C, and why
agenticloop 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 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- headroom:rtk-instructions --> How it starts
The opening of the file, as written. The whole thing — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Agentic Loop is the subject here, not the process
This repository authors the Agentic Loop methodology. AGENTIC_LOOP.md,
agents/, skills/, and backends/ are the product being maintained – they
are artifacts to edit, not a workflow to adopt for work in this repo.
Do not read AGENTIC_LOOP.md and then take on its roles, task-record
contract, worktree/parallel-lane rules, or event logging on your own
initiative. Ordinary work here – editing docs, fixing skills/backends,
answering questions, one-off changes – follows the rest of this AGENTS.md
directly, as normal manual repository work.
Only operate as Agentic Loop (adopt roles, create task records, spin up
worktrees/lanes, log events) when the user explicitly asks to run or dogfood
the methodology against this tree. Reading AGENTIC_LOOP.md to edit it or to
answer a question about it is fine and expected; adopting it as your process
is not.
Purpose
This file defines repository-wide rules for Agentic Loop itself.
Agentic Loop is a reusable, Markdown-first workflow toolkit for downstream repositories. It is maintained manually as normal repository work. Do not run Agentic Loop as an autonomous loop against itself.
Source of Truth
Use these files as the durable repository contract:
The docs/ directory is the user-facing setup and authoring surface. Keep it practical and adoption-oriented.
Stable Agentic Loop vocabulary belongs in AGENTIC_LOOP.md.
Optional private project context
When PROJECT_CONTEXT_ROOT is set and a request involves product direction,
architecture tradeoffs, prioritization, or development recommendations, read
PROJECT.md, STRATEGY.md, STATUS.md, and relevant decisions/ entries from
that directory before recommending a direction.
That context supplements but never overrides this public repository's source of truth. Surface conflicts explicitly, and never copy private planning or decision content into public files unless the user explicitly promotes it.
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.
- 2d ago First seen · 186 lines · 1,980 tokens per session scan C 4d655164ea76
agenticloop AGENTS.md is an instructions file published in the GitHub repository bartoszarendt/agenticloop (2 stars, last pushed 12d ago), licensed MIT. It adds 1,980 tokens to every session, about $0.0099 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
deer-flow AGENTS.md
AGENTS.md instructions for bytedance/deer-flow, covering agents.md, what is deerflow, service topology, repository map and commands: root vs. module.
deer-flow copilot-instructions.md
Copilot instructions for bytedance/deer-flow, covering copilot onboarding instructions for deerflow, 1) repository summary, 2) runtime and toolchain requirements, 3) build/test/lint/run - verified command sequences and a. bootstrap and install.
deer-flow CLAUDE.md
Claude Code instructions for bytedance/deer-flow: The repo's agent guidance lives in AGENTS.md so it is shared across coding agents (Claude Code, Codex, and others). Claude Code imports it below.
agents-universe CLAUDE.md
Instructions for agents-universe/agents-universe, covering agents universe — enterprise ai agent framework, project overview, monorepo layout, development commands and api (from packages/api/).
earnings-edge copilot-instructions.md
Copilot instructions for suhaas/earnings-edge, covering earnings-edge: agent development guidelines, architecture & core concepts, the graph, agent roles and tools & skills boundary.
earnings-edge AGENTS.md
AGENTS.md instructions for suhaas/earnings-edge, covering earnings-edge: agent development guidelines, architecture & core concepts, the graph, agent roles and tools & skills boundary.