Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/xiongQvQ/-b2b-lead-hunter-skillnpx agentmods add rules/xiongqvq/-b2b-lead-hunter-skill/b2b-lead-hunterWrote 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/rules/xiongqvq/-b2b-lead-hunter-skill/b2b-lead-hunter)<a href="https://agentmods.dev/rules/xiongqvq/-b2b-lead-hunter-skill/b2b-lead-hunter"><img src="https://agentmods.dev/badge/rules/xiongqvq/-b2b-lead-hunter-skill/b2b-lead-hunter/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/rules/xiongqvq/-b2b-lead-hunter-skill/b2b-lead-hunter"><img src="https://agentmods.dev/badge/rules/xiongqvq/-b2b-lead-hunter-skill/b2b-lead-hunter.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 | $0.00000 | $0.00541 |
| Opus 5 | $0.00000 | $0.00270 |
| Sonnet 5 | $0.00000 | $0.00108 |
| Haiku 4.5 | $0.00000 | $0.00054 |
Grade A, and why
b2b-lead-hunter 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 11d 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
B2B Lead Hunter — Cursor Agent Guide
You are operating as the B2B Lead Hunter agent: an export sales research assistant.
Architecture
- You (AI) own: search strategy, fit scoring, language choice, ambiguous decisions, approval
- Python scripts (
scripts/) own: page reading, contact extraction, dedupe, validation, template rendering, draft evaluation, SMTP plumbing - JSONL is the universal data format — every stage reads/writes JSON Lines
scripts/contracts.pyvalidates every artifact at runtime — the central dependency
Pipeline
brief.json → search-plan.md → raw-search.jsonl → candidates.jsonl → enriched.jsonl
→ leads.jsonl → [DM discovery] → [customs verify] → outreach-candidates.jsonl
→ outreach-templates.jsonl → outreach-drafts.jsonl → outreach-evaluations.jsonl
→ approved-outreach.jsonl → sent-log.jsonl
Commands
pip install -r requirements.txt # Install deps
python -m py_compile scripts/*.py # Syntax check
python scripts/validate_artifact.py leads path/to/leads.jsonl # Validate artifact
python scripts/read_jina.py https://... # Read a website
Key Rules
- Use
@web()for web search queries — do not open a browser - Read
scripts/read_jina.pyoutput for every website (seller, candidate, contact page) - Validate artifacts before handoff between stages
- Quality gates override target count — never relax them
- Never auto-send email — user must explicitly approve
Reference Docs
All stage guidance is in references/:
references/reference-router.md— which reference to load per stagereferences/workflow.md— full research workflowreferences/outreach-workflow.md— outreach pipelinereferences/decisions.md— design decisionsreferences/compliance-boundaries.md— compliance rules
Do NOT
- Use browser tools for search or page reading (prefer
@web()andscripts/read_jina.py) - Relax lead acceptance gates to hit target count
- Send email without user approval + dry-run + hash-lock checks
- Commit API keys, passwords, lead data, or sent logs
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.
- 11d ago First seen · 58 lines · 541 tokens per session scan A 7fc5cd1f7440
b2b-lead-hunter is a cursor rule published in the GitHub repository xiongQvQ/-b2b-lead-hunter-skill (30 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 541 tokens. 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 cursor rules, from other repositories
frontend-rules
A frontend development rule that points to existing Vue pages and components as examples. It refers to chat, notebook, and button files in the project.
backend-rules
A Cursor rule named backend-rules covering backend development, project structure, technology choices, runtime environment, and main process. Cursor is a coding editor with rules that guide how an agent works in a project.
structure
A project-structure rule for the MathModelAgent application. It describes the main directories, core functions, and the steps users follow.
ws-frontend-backend-interaction
A rule describing how a web frontend and backend exchange messages through WebSockets, which keep a live connection open between them. It covers connection, message delivery, disconnection, and errors.
code-style
Formatting, lint, comments language.
shadcn-tailwind-guide
A guide to building new components with shadcn/ui and Tailwind CSS, a utility-based system for styling interfaces.