Project Idea Validator — Discover What Already Exists Before You Build

You have access to the TinyFish CLI (tinyfish), a tool that runs browser automations from the terminal using natural language goals. This skill uses it to search GitHub and Dev.to in parallel, then synthesizes results into a gap analysis report.

Pre-flight Check (REQUIRED)

Before making any TinyFish call, always run BOTH checks:

1. CLI installed?

PowerShell:

Get-Command tinyfish; tinyfish --version

bash/zsh:

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:

Option 1 — CLI login (interactive):
tinyfish auth login
Option 2 — PowerShell (current session only):
$env:TINYFISH_API_KEY="your_api_key_here"
Option 3 — PowerShell (persist across sessions):
[System.Environment]::SetEnvironmentVariable("TINYFISH_API_KEY", "your_api_key_here", "User")
Then close and reopen PowerShell for it to take effect.

Option 4 — bash/zsh (Mac/Linux):
export TINYFISH_API_KEY="your_api_key_here"
Option 5 — Claude Code settings: Add to ~/.claude/settings.local.json:
{
  "env": {
    "TINYFISH_API_KEY": "your_api_key_here"
  }
}

Do NOT proceed until both checks pass.

What This Skill Does

Given a project idea (e.g. "a CLI tool that converts Figma designs to Tailwind components"), this skill:

Searches GitHub for existing repos with similar purpose, tech stack, or keywords
Searches Dev.to for articles, tutorials, or project showcases covering the same problem

It then synthesizes findings into a structured gap analysis: what exists, how mature it is, and where the opportunity still lives.

Core Command

tinyfish agent run --url <url> "<goal>"

Flags

Flag	Purpose
`--url <url>`	Target website URL
`--sync`	Wait for full result (no streaming)
`--async`	Submit and return immediately
`--pretty`	Human-readable formatted output

Step-by-Step Workflow

Step 1 — Search GitHub

Search for existing repositories matching the idea. Run with --sync since you need the full list before synthesizing.

tinyfish agent run --sync --url "https://github.com/search?q=<keywords>&type=repositories&s=stars&o=desc" \
  "Extract the top 10 search results as JSON: [{\"name\": str, \"owner\": str, \"description\": str, \"stars\": str, \"url\": str, \"last_updated\": str}]"

Example for a Figma-to-Tailwind CLI idea:

tinyfish agent run --sync \
  --url "https://github.com/search?q=figma+tailwind+cli&type=repositories&s=stars&o=desc" \
  "Extract the top 10 repositories as JSON: [{\"name\": str, \"owner\": str, \"description\": str, \"stars\": str, \"url\": str, \"last_updated\": str}]"

Step 2 — Search Dev.to

Search for articles and project posts covering the same problem space. Run in parallel with Step 1 results processing.

tinyfish agent run --sync --url "https://dev.to/search?q=<keywords>" \
  "Extract the top 10 articles as JSON: [{\"title\": str, \"author\": str, \"tags\": [str], \"published_at\": str, \"url\": str, \"reactions\": str}]"

Example:

tinyfish agent run --sync \
  --url "https://dev.to/search?q=figma+tailwind+component+generator" \
  "Extract the top 10 articles as JSON: [{\"title\": str, \"author\": str, \"tags\": [str], \"published_at\": str, \"url\": str, \"reactions\": str}]"

Parallel Execution

Steps 1 and 2 are independent — run them at the same time. Do NOT wait for GitHub before starting Dev.to.

Good — Parallel calls:

# Fire both simultaneously
tinyfish agent run --sync --url "https://github.com/search?q=<keywords>&type=repositories&s=stars&o=desc" \
  "Extract top 10 repositories as JSON: [{\"name\": str, \"owner\": str, \"description\": str, \"stars\": str, \"url\": str, \"last_updated\": str}]" &

tinyfish agent run --sync --url "https://dev.to/search?q=<keywords>" \
  "Extract top 10 articles as JSON: [{\"title\": str, \"author\": str, \"url\": str, \"reactions\": str}]" &

wait

Bad — Sequential calls:

# Don't do this — wastes time and gives the same results
tinyfish agent run --url "https://github.com/..." "...also search Dev.to..."

Each source is its own call. Always.

Step 3 — Synthesize Into a Gap Analysis

Once both sources return results, synthesize findings into this structure:

## Project Idea Validation: <idea title>

### What Already Exists
- <project/article> — <what it does, stars/reactions, last active>
- ...

### Maturity Assessment
- GitHub: <active / abandoned / fragmented>
- Dev.to coverage: <heavy / moderate / sparse>

### Gaps & Opportunities
- <specific gap #1>
- <specific gap #2>
- ...

### Verdict
<1–2 sentences: is the space crowded, open, or ripe for a better take?>

Use the raw JSON from both sources as input. Do not hallucinate repo names, star counts, or article titles — only use what TinyFish returned.

Keyword Strategy

The quality of results depends heavily on your search terms. Before running, derive 2–3 keyword variants from the idea:

Idea	Primary keywords	Variant keywords
Figma-to-Tailwind CLI	`figma tailwind cli`	`figma css export`, `design token tailwind`
AI code review bot	`ai code review github`	`llm pull request`, `automated code feedback`
Markdown-to-Notion sync	`markdown notion sync`	`notion import cli`, `notion api markdown`

Run separate parallel calls for each variant if the first pass returns sparse results.

Managing Runs

# List recent runs
tinyfish agent run list

# Get a specific run by ID
tinyfish agent run get <run_id>

# Cancel a running automation
tinyfish agent run cancel <run_id>

Output

The CLI streams data: {...} SSE lines by default. The final result is the event where type == "COMPLETE" and status == "COMPLETED" — the extracted data is in the resultJson field. Read the raw output directly; no script-side parsing is needed.

project-idea-validator