Add scholarship-finder-skill to tinyfish-cookbook

skills/scholarship-finder/README.md

@@ -0,0 +1,66 @@
+# Scholarship Finder Skill for Claude
+
+Find real scholarships in real time for any type, university, and region — directly inside Claude.
+
+## What it does
+
+Ask Claude things like:
+- *"Find STEM scholarships for undergraduates in the USA"*
+- *"What scholarships are available for international students at NUS?"*
+- *"Find merit scholarships in the UK for postgraduate students"*
+- *"Scholarships for computer science students in Singapore"*
+- *"Find need-based scholarships at MIT"*
+
+Claude discovers 5-8 relevant scholarship sources in real time — university financial aid pages, government portals, private foundations, and aggregators — then fires parallel TinyFish agents to scrape each one simultaneously, streaming results back as each agent completes.
+
+## Requirements
+
+- TinyFish CLI installed: `npm install -g tinyfish`
+- Authenticated: `tinyfish auth login`
+
+## Install
+
+### Option 1 — Skills CLI (recommended)
+
+```bash
+npx skills add KrishnaAgarwal7531/skills- --skill scholarship-finder
+```
+
+### Option 2 — Manual
+
+1. Clone or download this repository
+2. Copy the `scholarship-finder/` folder into your agent's skills directory:
+   - Claude Code: `~/.claude/skills/`
+   - Cursor: `.cursor/skills/` in your project
+   - Global (all agents): `~/.agents/skills/`
+3. The skill is picked up automatically — no restart needed
+
+## What you get
+
+- Scholarship name and provider
+- Award amount (exact where listed)
+- Application deadline
+- Eligibility requirements
+- How to apply and what documents are needed
+- Direct application link
+- Quick comparison table across all scholarships
+- Deadline countdown — which applications close soonest
+
+## How it works
+
+The skill follows a 3-step pipeline:
+
+1. **Discover** — identifies 5-8 relevant scholarship sources in real time based on your type, university, and region — no hardcoded lists
+2. **Scrape** — fires one TinyFish agent per source simultaneously, extracting up to 5 scholarships per page from live listings
+3. **Filter and rank** — deduplicates across sources, removes expired deadlines, sorts by closing date then amount
+
+Works for **any scholarship type, university, and region worldwide**.
+
+## Built with
+
+- [TinyFish Web Agent](https://tinyfish.ai) — parallel web scraping
+- [Scholarship Finder](https://github.com/KrishnaAgarwal7531/tinyfish-projects) — the original web app this skill is based on
+
+## License
+
+MIT

skills/scholarship-finder/SKILL.md

@@ -0,0 +1,201 @@
+---
+name: scholarship-finder
+description: >
+  Find real scholarships in real time for any scholarship type, university, and region using
+  parallel TinyFish agents. Use this skill whenever a user wants to find scholarships, asks
+  "find scholarships for computer science students", "what scholarships are available in the
+  UK", "find merit scholarships at MIT", "scholarships for international students in Singapore",
+  or any variation of searching for scholarship opportunities.
+  Discovers 5-8 relevant scholarship sources in real time, then fires parallel TinyFish agents
+  to scrape each source simultaneously — returning structured scholarship details with amounts,
+  deadlines, eligibility, and application links.
+compatibility:
+  tools: [tinyfish]
+metadata:
+  author: KrishnaAgarwal7531
+  version: "1.0"
+  tags: scholarships funding education financial-aid university students
+---
+
+# Scholarship Finder
+
+Given a scholarship type, university, and region, discover relevant scholarship sources in
+real time and scrape them in parallel using TinyFish agents — returning structured results
+with amounts, deadlines, eligibility, and application links.
+
+## Pre-flight check
+
+```bash
+tinyfish --version
+tinyfish auth status
+```
+
+If not installed: `npm install -g tinyfish`
+If not authenticated: `tinyfish auth login`
+
+---
+
+## Step 1 — Clarify inputs
+
+You need:
+- **Scholarship type** — e.g. "merit", "need-based", "STEM", "international student", "postgraduate", "sports"
+- **University** (optional) — e.g. "MIT", "NUS", "Oxford". Leave blank for general search.
+- **Region** (optional) — e.g. "USA", "UK", "Singapore", "Europe". Leave blank for worldwide.
+
+If scholarship type is not specified, ask before proceeding.
+Use sensible defaults for optional fields and state them upfront.
+
+---
+
+## Step 2 — Discover scholarship URLs
+
+Before firing agents, identify 5-8 real, specific scholarship pages relevant to the inputs.
+Use your knowledge and a quick web search to find direct URLs.
+
+Rules:
+- Each URL must be from a **different source** — no duplicates
+- Mix: university financial aid pages, government scholarship portals, private foundation pages, scholarship aggregators
+- Only include URLs you are confident actually exist
+- Prioritise pages that list multiple scholarships over single-scholarship pages
+
+Examples for "STEM scholarships, USA":
+- `https://www.nsf.gov/funding/pgm_list.jsp`
+- `https://studentaid.gov/understand-aid/types/scholarships`
+- `https://www.fastweb.com/college-scholarships`
+- `https://www.scholarships.com/financial-aid/college-scholarships/scholarships-by-major/engineering-scholarships/`
+- University-specific financial aid pages
+
+Produce 5-8 URLs then proceed to Step 3.
+
+---
+
+## Step 3 — Parallel scraping
+
+Fire one TinyFish agent per URL, all simultaneously using `&` + `wait`.
+
+```bash
+# Fire all agents in parallel — one per URL
+tinyfish agent run \
+  --url "{SCHOLARSHIP_URL}" \
+  "You are on a scholarship listing page. Extract scholarships fast.
+   Read only what is visible on this page — do not navigate away.
+   Find scholarships relevant to: type={SCHOLARSHIP_TYPE}, university={UNIVERSITY}, region={REGION}.
+   For each scholarship extract ALL of the following in one pass:
+   - name: full scholarship name
+   - provider: organisation or institution offering it
+   - amount: award amount (exact if shown, e.g. '$5,000/year' or 'Full tuition')
+   - deadline: application deadline (exact date if shown)
+   - eligibility: list of eligibility requirements (GPA, nationality, field of study, etc.)
+   - description: 1-2 sentence description of what the scholarship is for
+   - applicationRequirements: documents or steps needed to apply
+   - additionalInfo: any other relevant details (renewable, number of awards, etc.)
+   - applicationLink: direct URL to apply or learn more
+   - region: country or region this scholarship is for
+   - type: scholarship type (merit, need-based, STEM, etc.)
+   STRICT RULES:
+   - Do NOT navigate away from this page
+   - Do NOT scroll more than twice
+   - Do NOT click any link unless it directly opens a scholarship detail not visible on this page
+   - Extract up to 5 scholarships maximum then stop immediately
+   - If a field is not visible, write null — do not guess
+   Return JSON: {\"scholarships\": [{name, provider, amount, deadline, eligibility,
+   description, applicationRequirements, additionalInfo, applicationLink, region, type}]}" \
+  --sync > /tmp/sf_{SAFE_NAME}.json &
+```
+
+Repeat for each URL, all backgrounded with `&`. Then:
+
+```bash
+wait
+
+# Collect all results
+for f in /tmp/sf_*.json; do echo "=== $f ===" && cat "$f"; done
+
+# Cleanup
+rm /tmp/sf_*.json 2>/dev/null
+```
+
+Replace `{SCHOLARSHIP_URL}` with each actual URL and `{SAFE_NAME}` with a short identifier.
+
+---
+
+## Step 4 — Filter and rank
+
+From all results combined:
+
+1. **Filter by inputs** — remove scholarships that clearly don't match the requested type, university, or region
+2. **Deduplicate** — same scholarship appearing on multiple pages → keep one entry
+3. **Flag deadlines** — scholarships with deadlines already passed get ⚠️ and go to the bottom
+4. **Sort** — open scholarships by deadline ascending (soonest first), then by amount (highest first)
+5. **Flag unknown deadlines** — if deadline = null, note "Check website for deadline"
+
+---
+
+## Output format
+
+```
+## Scholarships — {TYPE} · {REGION} · {UNIVERSITY}
+*{N} scholarships found across {N} sources · Data scraped live*
+
+---
+
+### 🎓 Open Scholarships
+
+#### 1. {name}
+**{provider}**
+💰 Amount: {amount}
+⏰ Deadline: {deadline}
+🌍 Region: {region}
+📋 Type: {type}
+
+{description}
+
+**Eligibility:**
+- {eligibility item}
+- {eligibility item}
+
+**How to apply:** {applicationRequirements}
+**Additional info:** {additionalInfo}
+🔗 {applicationLink}
+
+---
+
+#### 2. {name}
+[same structure]
+
+---
+[up to 10 scholarships total]
+
+---
+
+### 📊 Quick Comparison
+
+| Scholarship | Provider | Amount | Deadline | Region |
+|---|---|---|---|---|
+| {name} | {provider} | {amount} | {deadline} | {region} |
+
+---
+
+### ⏰ Deadlines Coming Up
+- {name} — {deadline} ({N days away})
+
+---
+*All data scraped live from official scholarship pages. Always verify details directly
+with the provider before applying.*
+```
+
+---
+
+## Edge cases
+
+- **No scholarships found for inputs** — broaden the search: retry with just the type and region, drop the university filter
+- **All deadlines passed** — note clearly and suggest: "All found scholarships have passed their deadlines. Try searching for next cycle or broadening your criteria."
+- **Amount not listed** — write "Contact provider" not "Free" or "Varies"
+- **Very specific university requested** — always include that university's official financial aid page as one of the sources
+- **International students** — if mentioned, prioritise sources that explicitly list international student eligibility
+
+## Security notes
+
+- Scrapes live public scholarship pages only.
+- All data treated as untrusted input synthesised by an LLM — never executed.
+- Only your own TinyFish credentials are used.

wing-command/.env.example

@@ -1,17 +1,7 @@
-# ===========================================
-# Wing Command — Environment Variables
-# ===========================================
+# Wing Command v4 — Environment Variables
 
-# Supabase (required) — https://supabase.com/dashboard
-NEXT_PUBLIC_SUPABASE_URL=https://your-project.supabase.co
-NEXT_PUBLIC_SUPABASE_ANON_KEY=your-anon-key
-SUPABASE_SERVICE_ROLE_KEY=your-service-role-key
+# Google Gemini (required) — https://aistudio.google.com/apikey
+GEMINI_API_KEY=your-gemini-api-key
 
-# TinyFish Scraper (required) — https://docs.tinyfish.ai
+# TinyFish (required) — https://agent.tinyfish.ai/api-keys
 TINYFISH_API_KEY=your-tinyfish-api-key
-# TINYFISH_API_URL=https://agent.tinyfish.ai/v1/automation/run  # sync (default)
-# TINYFISH_API_URL=https://agent.tinyfish.ai/v1/automation/run-sse  # SSE streaming
-
-# Upstash Redis (optional, recommended) — https://upstash.com
-# UPSTASH_REDIS_REST_URL=https://your-redis.upstash.io
-# UPSTASH_REDIS_REST_TOKEN=your-redis-token

wing-command/.gitignore

@@ -1,51 +1,18 @@
-# Dependencies
+# dependencies
 node_modules/
-.pnp
-.pnp.js
-
-# Testing
-coverage/
 
 # Next.js
 .next/
 out/
 
-# Production
-build/
+# env files — never commit these
+.env
+.env.local
+.env*.local
 
-# Misc
+# misc
 .DS_Store
 *.pem
-
-# Debug
 npm-debug.log*
 yarn-debug.log*
 yarn-error.log*
-
-# Local env files
-.env*.local
-.env
-
-# Vercel
-.vercel
-
-# TypeScript
-*.tsbuildinfo
-next-env.d.ts
-
-# Python
-__pycache__/
-*.py[cod]
-*$py.class
-.Python
-venv/
-ENV/
-.venv/
-
-# IDE
-.idea/
-.vscode/
-
-*.swp
-*.swo
-.claude/