L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Données et recherche · SKILL.md

who-is-this

who-is-this

Research a person's public track record and give a short credibility assessment. Use only when the user explicitly invokes /who-is-this.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: who-is-this description: 'Research a person''s public track record and give a short credibility assessment. Use only when the user explicitly invokes /who-is-this.' disable-model-invocation: true triggers: [user, model]

who-is-this

Read the deepapi skill first. All search and scraping goes through DeepAPI. Never use built-in search, fetch, or a browser.

Seed

Accept a name, handle, URL, or profile screenshot. If missing, ask once; never guess a person.

Verify matching bio, company, location, and photo before deeper research. If identity is ambiguous, stop and ask.

Research

Find their X, LinkedIn, and GitHub first (dedicated endpoints, not site: search). Then run these in parallel:

  1. GitHub activity. POST /v1/scrape/github/profile with includeRepos: true, plus recent PRs (POST /v1/scrape/github/search, type: "pulls", query: "author:<username>", sort: "updated").
  2. Last 10 LinkedIn posts. POST /v1/scrape/linkedin/posts, maxItems: 10. Also scrape the LinkedIn profile if you have the slug.
  3. Deep research. POST /v1/research/deep on who they are, what they have actually done, and whether the public story holds.
  4. 50 newest X posts. POST /v1/scrape/twitter/search with the handle, sort: "latest", maxItems: 50. Also pull POST /v1/scrape/twitter/user.

Follow DeepAPI polling. If a platform is missing or private, say so. Do not invent a profile.

What to extract

Keep only the 3 facts that best explain who the person is; omit the rest.

  • Real track record beats bio. Jobs, products, exits, code, talks.
  • Self-reported numbers stay labeled "their claim". Verified numbers say "verified".
  • Include posting topics only if they change the verdict.
  • Choose the archetype their track record supports: builder, marketer, operator, researcher, grifter, investor, recruiter, hobbyist, etc.

Ignore congrats, logo spam, paid "king of X" press, and follower-count flexing.

Output

Maximum 100 words after the header. Use short, plain-English sentences. No tables, sub-bullets, tweet dumps, or research narration.

**Full name** — [@handle](https://x.com/handle) · [LinkedIn](url) · [GitHub](url) · [Site](url)

**Who:** One sentence. Where they are and what they do now.

**Track record:** Max 3 bullets. Only what explains who they are. Dates. Label claims vs verified.

**Verdict:** One line. The archetype, whether the story holds, and why.

Fictional example:

**Jane Doe** — [@janedoe](https://x.com/janedoe) · [LinkedIn](https://linkedin.com/in/janedoe) · [GitHub](https://github.com/janedoe)

**Who:** Berlin solo founder of Acme, an open-source Postgres proxy.

**Track record:**
- 2024–now: Acme founder. 2.1K stars (verified); no revenue (her claim).
- 2019–2024: Backend engineer at Zalando and N26.
- Claims "50K users"; unverified.

**Verdict:** Solo builder with real code; user numbers remain unverified.

Omit a missing profile link instead of faking it.

Follow-ups: 1–3 sentences. Re-scrape only if the first pass missed that platform.

Skills enregistrées