AutoHelix is a harness that lets an AI agent repeatedly propose changes toward a measurable goal while checks decide which changes are accepted. It is for tasks such as code optimization or model training where constraints, metrics, file scope, isolation, and budgets need to be controlled. Its catalogue contains instructions and a skill for running the improvement loop.
Borrowing it
Nothing to install: this file belongs to awslabs/AutoHelix. 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/awslabs/AutoHelix/main/CLAUDE.mdgit clone --depth 1 https://github.com/awslabs/AutoHelixWrote 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/awslabs/autohelix/claude-md)<a href="https://agentmods.dev/instructions/awslabs/autohelix/claude-md"><img src="https://agentmods.dev/badge/instructions/awslabs/autohelix/claude-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/awslabs/autohelix/claude-md"><img src="https://agentmods.dev/badge/instructions/awslabs/autohelix/claude-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 | $0.01370 | $0.01370 |
| Opus 5.5 | $0.00548 | $0.00548 |
| Sonnet 5.5 | $0.00274 | $0.00274 |
| Haiku 4.5 | $0.00137 | $0.00137 |
Grade A, and why
AutoHelix CLAUDE.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 1mo 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.
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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AutoHelix Development
AutoHelix is an agent-powered iterative optimization tool. It runs AI agents in a loop — each iteration is isolated in a git worktree, validated, and only merged when it passes. See README.md for user-facing docs. For setting up training/eval (ML benchmark) cells, see docs/ml-experiments.md.
Setup
Activate the virtual environment before running any commands:
source .venv/bin/activate
If the venv doesn't exist yet:
uv venv
source .venv/bin/activate
uv pip install -e ".[dev]"
Project Layout
src/autohelix/harness.py- main iteration loop (start here)src/autohelix/cli.py- CLI commands (init, run, clear, watch, report)src/autohelix/config.py- configuration dataclasses and parsingsrc/autohelix/agents/- agent backends (claudecode, codex, opencode, mock)src/autohelix/sandbox.py- git worktree management and scope enforcementsrc/autohelix/checks.py- constraint and metric executionsrc/autohelix/history.py- iteration history (JSONL)src/autohelix/display.py- live Rich displaysrc/autohelix/prompt_template.py- agent prompt renderingtests/- test suite (pytest)examples/- example projects (sorting, ml-recipe, writing, task-queue, research, workflow-optimization, algotune, posttrain, kernelbench, nested-autohelix)scripts/run_dev_test.sh- quick real-agent test against a bundled example (supports--parallel N)docs/- user-facing documentation (plain markdown)
CLI Commands
autohelix init # Initialize project (creates .autohelix/ and config template)
autohelix run # Start or resume optimization loop
autohelix parallel # Run multiple workers in parallel on the same goal
autohelix clear # Archive state, start fresh
autohelix watch # Live-tail agent output (second terminal)
autohelix hint # Send a hint to the running agent (picked up next iteration)
autohelix report # Agent-written analysis of the run
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.
- 1mo ago Changed · +2 lines · +13 tokens per session ffbe5d1b43c1
- 1mo ago First seen · 81 lines · 1,357 tokens per session scan A e854c9409d96
AutoHelix CLAUDE.md is an instructions file published in the GitHub repository awslabs/AutoHelix (32 stars, last pushed today), licensed Apache-2.0. It adds 1,370 tokens to every session, about $0.0055 per session on Opus 5.5. 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 2026-08-30.
Other instructions, from other repositories
runtm copilot-instructions.md
Copilot instructions for runtm-ai/runtm, covering runtm tool instructions, critical rules, structure, adding endpoints and app/api/v1/myendpoint.py.
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.