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/mingye-lu/agenticcrawler/agents-mdgit clone --depth 1 https://github.com/Mingye-Lu/AgenticCrawlerWhat 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 | $0.05394 | $0.05394 |
| Opus 5 | $0.02697 | $0.02697 |
| Sonnet 5 | $0.01079 | $0.01079 |
| Haiku 4.5 | $0.00539 | $0.00539 |
Grade A, and why
AgenticCrawler 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 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.
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 — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Project
acrawl is a native-Rust LLM-driven web crawler. A user provides a natural-language goal; the agent plans, navigates, and extracts structured data via a 42-tool toolbox (31 browser + 4 agent-control + 7 script). It ships as a single binary with three modes: an interactive Ratatui TUI REPL (requires a TTY), non-interactive prompt (one-shot) / --resume (slash-command replay), and mcp (built-in MCP server over stdio).
Commands
cargo build --release # produce ./target/release/acrawl
cargo test --workspace # run full test suite (~1,100 tests)
cargo test -p <crate> <test_name> # run a single test (e.g. -p agent mvp_tool_specs_contains_expected_42_tools)
cargo clippy --workspace --all-targets -- -D warnings # lints must be clean (workspace lints set pedantic = warn)
cargo fmt --check # format check
./target/release/acrawl # launch REPL
./target/release/acrawl prompt "scrape all titles from example.com" # one-shot
./target/release/acrawl mcp # launch MCP server (stdio)
./target/release/acrawl mcp install # interactive IDE installer
./target/release/acrawl --resume session.json /status /compact # non-interactive session maintenance
# Non-interactive provider setup (agent/CI use)
acrawl auth anthropic --api-key "sk-ant-..." # configure credentials
acrawl auth openai --api-key "sk-..." # other providers same pattern
acrawl auth amazon-bedrock --access-key AKIA... --secret-key ... --region us-east-1
acrawl config set model anthropic/claude-sonnet-4-6 # set default model
acrawl auth status --check anthropic # gate: exit 0 if ready, 3 if not
acrawl auth status # show all configured providers
acrawl auth list # list all available providers
acrawl config get headless # read a setting
acrawl config set headless false # write a setting
acrawl mcp install --client opencode # install MCP for one IDE
acrawl mcp install --all --yes # install for all IDEs
# Exit codes: 0=ok 1=error 2=usage/config 3=not-configured
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 · 230 lines · 5,394 tokens per session scan A b4413cc96379
AgenticCrawler AGENTS.md is an instructions file published in the GitHub repository Mingye-Lu/AgenticCrawler (12 stars, last pushed 1mo ago), licensed MIT. It adds 5,394 tokens to every session, about $0.0270 per session on Opus 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
cloakbrowser-mcp AGENTS.md
Instructions for swimmwatch/cloakbrowser-mcp, covering agents.md, about the project, golden rules, project layout and daily commands.
browser-use CLAUDE.md
Claude Code instructions for browser-use/browser-use, covering claude.md, high-level architecture, core components, event-driven browser management and cdp integration.
obscura AGENTS.md
Instructions for h4ckf0r0day/obscura, covering agents.md, build, rendering and stealth, no rendering, with rustls or stealth and test.
browser-harness AGENTS.md
Instructions for browser-use/browser-harness, covering code priorities and contributing.
browser-control AGENTS.md
Instructions for keon/browser-control: Use this project as a tiny browser pipe, not an agent framework.
stealth-browser-mcp AGENTS.md
Instructions for vibheksoni/stealth-browser-mcp, covering agent instructions, start here, recommended proxy services, environment and validation.