/stalk-my-interviewer

name: stalk-my-interviewer
description: Research an interviewer online before a meeting using parallel TinyFish agents and return a structured prep report. Use this skill when a user says "research my interviewer", "I have an interview with [name] at [company]", "stalk my interviewer", "find out about [person] before my interview", "who is my interviewer", "prepare for interview with [name]", "look up my interviewer", or any request to learn about a specific person before meeting them professionally.

Stalk My Interviewer

Deploy parallel TinyFish agents to research an interviewer across LinkedIn, GitHub, Twitter/X, news, and conference platforms — then synthesize a structured prep report so you walk in knowing exactly who you're talking to.

Pre-flight Check (REQUIRED)

Before making any TinyFish call, always run BOTH checks:

**1. CLI installed?**

which tinyfish && tinyfish --version || echo "TINYFISH_CLI_NOT_INSTALLED"

If not installed, stop and tell the user: > Install the TinyFish CLI: `npm install -g @tiny-fish/cli`

**2. Authenticated?**

tinyfish auth status

If not authenticated, stop and tell the user: > You need a TinyFish API key. Get one at: https://agent.tinyfish.ai/api-keys > > Then authenticate: > ``` > tinyfish auth login > ```

Do NOT proceed until both checks pass.

---

Step 1 — Gather inputs

You need:

• **Interviewer's full name** — e.g. "Sarah Chen"
• **Company** — e.g. "Stripe", "Anthropic", "Linear"
• **Role you're interviewing for** (optional but improves output) — e.g. "Senior Software Engineer"

If any are missing, ask before proceeding. If the name is very common (e.g. "John Smith"), ask for company and role to disambiguate before searching.

---

Step 2 — Parallel research

Fire all agents simultaneously. Every agent searches a different surface — run them all at once using `&` + `wait`.

# Agent 1 — LinkedIn
tinyfish agent run \
  --url "https://www.linkedin.com/search/results/people/?keywords={FULL_NAME_ENCODED}+{COMPANY_ENCODED}" \
  "You are on a LinkedIn people search results page. Find the profile for {FULL_NAME} who works or worked at {COMPANY}.
   Click the most relevant result.
   On their profile extract:
   - Current job title and company
   - Previous roles (last 3 positions: title, company, duration)
   - Education (degrees, institutions)
   - Skills listed (top 10)
   - Summary / About section (if visible)
   - How long they have been at {COMPANY}
   STRICT RULES:
   - Click only the most relevant profile result — do not browse multiple profiles
   - Do NOT scroll more than twice on the profile page
   - If the page asks you to log in, extract whatever is visible before the gate and return it
   - Do NOT click any other links
   Return JSON: {name, current_title, current_company, tenure_at_company, previous_roles: [{title, company, duration}], education: [{degree, institution}], skills: [], summary}" \
  --sync > /tmp/smi_linkedin.json &

# Agent 2 — GitHub (relevant if role is technical)
tinyfish agent run \
  --url "https://github.com/search?q={FULL_NAME_ENCODED}+{COMPANY_ENCODED}&type=users" \
  "You are on GitHub user search results for {FULL_NAME} at {COMPANY}.
   Find the most likely profile match. Click it.
   On their GitHub profile extract:
   - Username
   - Bio
   - Location
   - Company listed on profile
   - Pinned repositories (name, description, language, stars)
   - Most used programming languages (visible in stats or repos)
   - Any notable open source contributions or projects
   STRICT RULES:
   - Click only the single most relevant result
   - Do NOT navigate to individual repos
   - Read only what is visible on their profile page
   - If no clear match found, return {found: false}
   Return JSON: {found: bool, username, bio, location, pinned_repos: [{name, description, language, stars}], languages: [], notable_work}" \
  --sync > /tmp/smi_github.json &

# Agent 3 — Twitter/X
tinyfish agent run \
  --url "https://x.com/search?q={FULL_NAME_ENCODED}+{COMPANY_ENCODED}&src=typed_query&f=user" \
  "You are on Twitter/X user search results for {FULL_NAME} at {COMPANY}.
   Find the most likely profile match. Click it.
   On their Twitter profile extract:
   - Display name and handle
   - Bio
   - Pinned tweet (if any)
   - Topics they tweet about most (infer from visible tweets — read up to 10)
   - Any strong opinions or recurring themes
   - Approximate tweet frequency / activity level
   STRICT RULES:
   - Click only the most relevant profile
   - Read only the first 10 visible tweets — do NOT scroll further
   - Do NOT click any tweet links or replies
   - If no match found, return {found: false}
   Return JSON: {found: bool, handle, bio, pinned_tweet, topics: [], opinions: [], activity_level}" \
  --sync > /tmp/smi_twitter.json &

# Agent 4 — Google News & web mentions
tinyfish agent run \
  --url "https://www.google.com/search?q=\"{FULL_NAME_ENCODED}\"+\"{COMPANY_ENCODED}\"&tbm=nws" \
  "You are on Google News search results for {FULL_NAME} at {COMPANY}.
   Read the titles and snippets of the first 10 visible news results.
   Extract:
   - Any articles authored by or quoting {FULL_NAME}
   - Key topics they are associated with in the news
   - Any notable achievements, announcements, or controversies mentioned
   STRICT RULES:
   - Do NOT click any article links
   - Read only titles and snippets visible in the search listing
   - Maximum 10 results then stop
   Return JSON: {mentions: [{title, snippet, source, date}], topics: [], authored_articles: []}" \
  --sync > /tmp/smi_news.json &

# Agent 5 — Company engineering blog
tinyfish agent run \
  --url "https://www.google.com/search?q=site:{COMPANY_DOMAIN}+\"{FULL_NAME_ENCODED}\"" \
  "You are on Google search results filtered to {COMPANY}'s website for content authored by or mentioning {FULL_NAME}.
   Read the visible results.
   Extract:
   - Any blog posts, articles, or pages authored by {FULL_NAME}
   - Topics they write