Needs radar
Skill mohamed-amine-ben-mallessa/needs-radar-mcp/skills/needs-radar
Detect company needs and generate ranked B2B prospect lists from job postings (reverse prospecting) via the needs-radar MCP server. Use when the user wants to find prospects, see who is hiring in an area/sector, do B2B prospecting, identify potential clients/leads, or analyze a territory's hiring market. A company that is hiring has a project, a budget, and adjacent unmet needs. Keywords: prospect, prospecting, lead, B2B, who is hiring, potential clients, sales opportunity, hiring signals, company needs.From its SKILL.md
npx -y skills add mohamed-amine-ben-mallessa/needs-radar-mcp --skill needs-radarAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its file declares
Copied from the file, not written here
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
2.8 KB, as published. Nobody here has run it
Needs Radar — reverse B2B prospecting from job postings
A job posting is a buying signal. Use the needs-radar MCP server to turn postings into ranked prospects. Respond in the user's language.
Tools
| Tool | Use |
|---|---|
scan_besoins | scan openings for a zone/sector/role → ranked prospect cards (+ staffing agencies separated, + market aggregations) |
market_stats | market tension for a role/zone (volume, contract mix, experience demand) |
scan_besoins params: motsCles (need/role), departement (e.g. "69"), commune (INSEE), region, distance, codeROME, maxOffres (10–150, default 100).
Score = openings×3 + company-size weight (SMEs favored) + recency. signalRecent: true = hot (posting < 30 days).
Full prospecting flow (compose with other tools the agent has)
The server returns the raw signal. To produce a real prospect brief, chain:
scan_besoins→ companies hiring in the target.- Enrich each promising one: company registry (search by name → size, age, status, leadership), open-data (public grants/tenders = budget signals), web/search/social (news, funding, decision-maker).
- LLM analysis: the real underlying need · why it's a fit · what to pitch · a drafted opener. A fast/cheap model is enough; fall back across providers to avoid rate limits.
- Deliverable: prospect cards (company · need · why · offer · drafted message). Offer to save to a notes vault or export CSV.
Variants
- By zone:
scan_besoins(departement="69")with no keyword → everyone hiring in Lyon. - By need:
scan_besoins(motsCles="site web")→ companies with a web need. - Multi-source: cross with other job sources for companies absent from France Travail; dedupe by name.
Notes & ethics
- France only (France Travail source). No SIRET in offers → registry matching is by name (imperfect).
- Staffing agencies are filtered by NAF; consulting/recruitment firms may still appear — flag them in analysis.
- Scoring is a transparent heuristic, not ground truth. Tune for the ideal customer.
- Legitimate, personalized B2B outreach only. Never spam.
Setup
Needs FT_CLIENT_ID / FT_CLIENT_SECRET (free France Travail app subscribed to "Offres d'emploi v2"). See CONFIGURATION.md.