By Parcha-ai
Deep research superpowers for your AI agent. Three research tiers (quick ~25s, deep ~5min, ultra up to 1hr) powered by GREP — the #1 deep research engine.
Build interactive HTML web apps via Grep — websites, dashboards, data explorers, calculators, comparison tools, charts, custom UIs. Use when the user asks for an interactive app, "build me a tool that does X", "make an interactive Y", or anything requiring a runnable HTML/JS deliverable beyond a static report. Routes to the app-builder expert with effort=build (~$2.00, 10-15 min). Returns an index.html in the job workspace; the skill prints the file URL when ready.
Create research-backed slidedecks via Grep. Use when the user asks for a presentation, pitch deck, slides, "deck about X", or "presentation on Y". Covers market research, competitive analysis, product pitches, technical overviews, sales enablement. Returns an HTML slidedeck (slides.html or deck.html) in the job workspace. Routes to the app-builder expert with output_type=slidedeck — v2 translates that to expert_id=app-builder + effort=build + a "Create a slidedeck about: " question prefix (~$2.00, 10-15 min).
Build comparison spreadsheets via Grep with structured tabular data — vendor comparisons, market data tables, financial models, dataset analyses, competitive matrices, pricing tables. Use when the user asks for a spreadsheet, table, CSV, comparison matrix, or comparative data. Returns an HTML spreadsheet (with copyable cells / export-CSV button) in the job workspace. Routes to the app-builder expert with output_type=spreadsheet — v2 translates to expert_id=app-builder + effort=build + a "Build a spreadsheet of: " prefix (~$2.00, 10-15 min).
Continue an existing Grep research job with a follow-up question, building on the prior job's findings without re-running the entire investigation. Use when the user says "follow up on that research", "continue the X investigation", "go deeper on the third finding", or references a previous job by id/slug ("continue grep.ai/research/foo-bar-baz with: ...").
Route research to the right Grep public domain expert (legal, medical, patent, financial, real estate, supply chain, maritime, etc.). Use when the user mentions a specialised domain ("legal research", "patent landscape", "biotech clinical trials", "real estate due diligence", "supply chain mapping") OR says "use the X expert". The skill reads resources/experts.md (snapshot of GET /api/v2/experts), picks the best match by keyword + sample-question similarity, and submits a research job with explicit expert_id.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Give your AI agent deep research superpowers. GREP Research Skills connects Claude Code, Cowork, and OpenClaw to GREP — the #1 deep research engine.
The wallet/PAYG gateway path (Stripe Link, Base USDC,
Receipt pi_xxxauth, x402 funding) lives on thempp-gateway-futurebranch. It returns once the backend's MPP gateway is GA. Until then this repo targets the v2 API only — Descope JWT orgrp_*API key.
npx grep-research-skills
That's it. Works with Claude Code, Cowork, and OpenClaw — the installer auto-detects your environment.
Requirements: Node.js 18+
For Cowork and Claude.ai, GREP is packaged as a single /research skill (instead of 8 individual skills). One router skill handles everything — it reads the right reference file based on your intent.
grep-research-skills-v0.2.0.zip/research skill appears in your org — all team members get accessWhy one skill? Claude Code discovers skills as individual symlinks, so 16 skills = better auto-triggering. Cowork/Claude.ai loads skills from a zip, where a single well-described router skill with reference files is more practical and avoids cluttering the skill list.
Building the zip from source:
npm install # one-time, pulls archiver (the only devDependency)
npm run build:cowork # → dist/grep-research-skills-v0.2.0.zip
Pure-JS — no system zip binary required. Works on macOS, Linux, and Windows.
Required: Allow network access. In your Cowork org settings under Code execution → Allow network egress, add these domains to the allowlist:
api.grep.ai
preview-api.grep.ai
api.descope.com
(Add preview-api.grep.ai only if you're testing against the preview deployment via GREP_API_BASE.)
Without this, the sandbox can't reach the GREP API or authenticate via Descope.
/install-plugin parcha-ai/grep-research-skills
git clone https://github.com/parcha-ai/grep-research-skills.git ~/.grep-research-skills && ~/.grep-research-skills/setup
| Skill | Triggers on | What it does |
|---|---|---|
/grep-login | "log in to grep", "grep auth" | Authenticate via email OTP or API key |
/grep-status | "grep status", "what plan am I on" | Account status + recent jobs |
/grep-upgrade | "upgrade grep", "buy more credits" | Choose / change subscription plan (Free / Pro / Ultra / PAYG) |
/grep-skill-creator | "make a grep skill" | Create a new SKILL.md for any agent skill, powered by deep research |
/grep-plan | "plan a grep research", "research-informed plan" | Research best practices + your codebase context before you /plan |
/quick-research | "fast research / lookup" | ~25s sourced one-liner — version checks, API endpoints, quick lookups |
/research | "research X" (default tier) | ~5 min comprehensive report with citations — the default for most tasks |
/ultra-research | "deep / exhaustive research on X" | Up to 1hr investigation — security audits, legal, ecosystem surveys |
/grep-mcp | "install grep MCP", "grep as MCP server" | Wire Grep into .mcp.json as 4 native MCP tools |
/grep-domain-expert | "use the legal/medical/patent/... expert" | Route to one of 27 public domain experts (legal, medical, financial, real estate, supply chain, maritime, etc.) |
/grep-build-app | "build me an interactive app for X" | Interactive HTML web apps via the app-builder expert (effort=build, ~$2, 10-15min) |
/grep-build-slidedeck | "make me a slidedeck about X" | Research-backed HTML slidedeck with arrow-key nav + PDF export |
/grep-build-spreadsheet | "build a spreadsheet of X" | Sortable HTML spreadsheet with CSV export |
/grep-research-workflow | "investigate X and then make Y" | Multi-step chain: orient → deep dive → optional build artifact |
/grep-with-context | "research using these PDFs" | Upload files as research inputs (PDFs, CSVs, images) |
/grep-continue | "follow up on that research" | Continue an existing job with a new question, inheriting prior research context |
Bold rows are new in 0.2.0.
| Skill | Description |
|---|---|
/research | All features in one skill — routes to the right workflow (deep, quick, ultra, plan, skill-creator, login, upgrade, status) based on what you ask for |
npx claudepluginhub parcha-ai/grep-research-skills --plugin grep-research-skillsUltra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Frontend design skill for UI/UX implementation
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Memory compression system for Claude Code - persist context across sessions
Marketing skills for AI agents — conversion optimization, copywriting, SEO, paid ads, ad creative, and growth
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.