Borrowing it
Nothing to install: this file belongs to mdsohaib/screenshot-time-machine. 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/mdsohaib/screenshot-time-machine/main/AGENTS.mdgit clone --depth 1 https://github.com/mdsohaib/screenshot-time-machineWrote 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/mdsohaib/screenshot-time-machine/agents-md)<a href="https://agentmods.dev/instructions/mdsohaib/screenshot-time-machine/agents-md"><img src="https://agentmods.dev/badge/instructions/mdsohaib/screenshot-time-machine/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/mdsohaib/screenshot-time-machine/agents-md"><img src="https://agentmods.dev/badge/instructions/mdsohaib/screenshot-time-machine/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 | $0.00979 | $0.00979 |
| Opus 5.5 | $0.00392 | $0.00392 |
| Sonnet 5.5 | $0.00196 | $0.00196 |
| Haiku 4.5 | $0.00098 | $0.00098 |
Grade A, and why
screenshot-time-machine 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 2mo 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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md: how to work in this repo (humans and AI agents)
screenshot-time-machine (stm) is a small Node CLI: find the localhost app → discover pages → full-page screenshot each → dated folder + manifest.json → report which pages changed. Thousands of people will read this source. Optimize for readability, one runtime dependency, and excellent error messages.
Setup (5 commands)
git clone https://github.com/mdsohaib/screenshot-time-machine && cd screenshot-time-machine
npm ci
npm test # node:test: unit + CLI + e2e against the built-in fixture site
npm run fixture # keeps the fixture site up on 127.0.0.1:<port> for manual runs
node bin/stm.js --url http://127.0.0.1:<port>
npm run check = TypeScript over JSDoc (checkJs, strict). npm run lint = Biome. Both run in CI.
Module map (v0.1)
| File | Owns |
|---|---|
bin/stm.js |
Node-version guard, EPIPE guard, process.exitCode = await main() |
src/cli.js |
util.parseArgs, subcommand dispatch, human summary strings |
src/index.js |
run(options) / list(options), the programmatic API the tests call |
src/detect.js |
localhost port probing, "several apps" resolution |
src/discover.js |
URL normalization, link filtering, sitemap parsing (regex, no XML dep) |
src/capture.js |
browser bootstrap (cached shell → chrome → msedge → install), per-page pipeline, worker pool, overlay handling |
src/store.js |
snapshot folder naming, slug, incremental manifest, sha256 "changed", gitignore append |
src/list.js |
stm list |
src/types.d.ts |
Manifest, PageResult, RunOptions |
skills/stm/SKILL.md |
the agent skill (stm skill prints it) |
test/fixture/site.js |
the one fixture site (unit, e2e, CI smoke, GIF, benchmark) |
Hard rules (a PR that breaks one will be closed)
- One runtime dependency (
playwright-core, exact-pinned). No new dependency for XML, colors, args, tables, spinners. Useutil.parseArgs,util.styleText, regex. - No new flags or commands without an issue first. The answer is usually no. A good-first-issue never adds a flag.
- stdout is only the summary (human) or one JSON document (
--json). Progress goes to stderr, and only when stderr is a TTY.--jsonprints ≤ 2 stderr lines. - A failing page never aborts a run. Record it, continue, exit 2.
- stm never deletes anything. It writes only inside
--outand may append one line to an existing.gitignore(announced). - No page-derived text (titles, headings, body) in
--jsonor the manifest. That output is read by AI agents, so it is a prompt-injection channel. Errors are truncated to 200 chars. - No config file, no daemon/watch mode, no telemetry, no hooks or MCP servers in the plugin.
- Readability beats line count. Never golf. Boring over clever.
execFileonly, nevershell: true. - Vocabulary in user-facing strings: a folder is a "snapshot", a PNG is a "screenshot", URLs are "pages", the target is "your app". Never "routes", "dev server", "timeout" in human mode.
- You must understand your change. Using AI tools is fine. Submitting generated output you haven't reviewed and can't explain is not. Say in the PR if an agent wrote most of 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.
- 2mo ago First seen · 53 lines · 979 tokens per session scan A 1ec2df5fe7b3
screenshot-time-machine AGENTS.md is an instructions file published in the GitHub repository mdsohaib/screenshot-time-machine (4 stars, last pushed 5d ago), licensed MIT. It adds 979 tokens to every session, about $0.0039 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 2026-08-31.
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).
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.