Borrowing it
Nothing to install: this file belongs to haseab/retrace. 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/haseab/retrace/main/AGENTS.mdgit clone --depth 1 https://github.com/haseab/retraceWrote 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/haseab/retrace/agents-md)<a href="https://agentmods.dev/instructions/haseab/retrace/agents-md"><img src="https://agentmods.dev/badge/instructions/haseab/retrace/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/haseab/retrace/agents-md"><img src="https://agentmods.dev/badge/instructions/haseab/retrace/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 | $0.05833 | $0.05833 |
| Opus 5.5 | $0.02333 | $0.02333 |
| Sonnet 5.5 | $0.01167 | $0.01167 |
| Haiku 4.5 | $0.00583 | $0.00583 |
Grade C, and why
retrace 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.
rm -rf .build/ How it starts
The opening of the file, as written. The whole thing — 501 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Retrace - Agent Guide
Standard: This file follows the AGENTS.md specification - a vendor-agnostic standard for AI agent guidance. For human-readable project information, see README.md.
Retrace is a local-first screen recording and search application for macOS, inspired by Rewind AI. It captures screens, extracts text via OCR, and makes everything searchable—all locally on-device.
Status: Core screen capture (CGWindowListCapture), OCR (Vision), full-text search (FTS5), HEVC encoding, and Rewind import are working. Audio transcription and vector search are planned for future releases.
Quick Reference
- Module-Specific Instructions: Each module has its own
AGENTS.mdfile in its directory - Human Documentation: README.md, CONTRIBUTING.md, and AI_ISSUE_TEMPLATE.md
- Issue Reporting: Use
AI_ISSUE_TEMPLATE.mdandgh issue create --body-file ...for AI-authored GitHub issues - Bug Fixes by Non-Owners: If the user is fixing a bug/crash and does not appear to be the repo owner, encourage them to create or link a GitHub issue before making code changes
- Technical Audit Docs:
local/docs/(includes deep-dive implementation and performance audit notes)
Project Commands
Cloudflare Accounts
- Use the installed account wrappers:
wrangler personal ...andwrangler work .... Do not use barewranglercommands. wrangler personalselects the personal Cloudflare account ([email protected]); use it for Retrace'sretraceR2 bucket and release DMG uploads.wrangler workselects the work Cloudflare account ([email protected]); use it for work resources, not Retrace's release bucket.- Verify identity and resource access with the selected wrapper, for example
wrangler personal whoamiandwrangler personal r2 bucket list. A successful identity check alone does not prove R2 permission. - Upload releases with
wrangler personal r2 object put retrace/Retrace-v<VERSION>.dmg --file=<DMG_PATH> --remote, then verify the public artifact before deploying the appcast.
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 · +11 lines · +266 tokens per session 7e2e16270a00
- 2mo ago First seen · 490 lines · 5,567 tokens per session scan C 8750a1c90aed
retrace AGENTS.md is an instructions file published in the GitHub repository haseab/retrace (166 stars, last pushed yesterday), licensed MIT. It adds 5,833 tokens to every session, about $0.0233 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.
retrace AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| retrace AGENTS.mdhaseab/retrace | Instructions file | 166 | yesterday | MIT | C | 5,833 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,516 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| vscode model-prompts.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 2,022 | GitHub Copilot |