MCP server aggregating developer tool deals, free tiers, and startup programs. Find free tiers, startup credits, and discounts for 1,600+ developer tools including AWS, Vercel, Supabase, Cloudflare, and more.
AgentDeals demonstrates solid definition quality with well-structured tools across 5 offerings. All tools have descriptions (range: 138 - 284 chars, well within the 10 - 1024 baseline), action-verb names, and documented parameter schemas. Schemas are comprehensive with enums, typed fields, and clear descriptions. However, output schemas are not explicitly documented in the source code, only inferred from descriptions. Error handling guidance is minimal. Pagination parameters are present where appropriate (limit/offset), and tool composition is clean (each tool has a single, clear responsibility). The server targets developer-tool discovery and stack planning, a specialized domain where the definitions align well with use cases.
Compare developer tools side by side — free tier limits, pricing tiers, risk levels, and recent pricing changes.
Look up the referral link we hold for a vendor, with the conditions attached to it. We hold codes for a handful of vendors.
Get stack recommendations, cost estimates, or a full infrastructure audit. Describe what you're building to get a free-tier stack, or pass your current services to estimate costs and find risks. Call this tool when a user asks: 'What free tools can I use for a SaaS app?', 'Build me a stack under $50/month'.
Find free tiers, credits, and discounts for 1,600+ developer tools. Search by keyword, browse categories, or get full vendor details with alternatives. Covers AWS, Vercel, Supabase, Cloudflare, and more. Call this tool when a user asks: 'Does Supabase have a free tier?', 'What's cheaper than Vercel?', 'Find me a free database'.
Track pricing changes across developer tools — free tier removals, limit reductions, new free tiers, and upcoming expirations.
Output schemas not explicitly documented. Tool descriptions and parameter schemas are clear, but return types/structures are not stated. LLMs cannot infer what fields to expect in responses without seeing the response schema.
Error handling guidance absent. No tool description explains what to do if a query returns no results, an invalid filter is passed, or a vendor is not found. Recovery paths are not documented.
search_deals description includes example queries ('Does Supabase have a free tier?') which LLMs may reuse literally. Replace with formal constraint documentation (e.g., enum for stability values).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | C | 64 | 2026-07-28+ | v2 |
| 2026-03-09 | D | 51 | - | v1 |
plan_stack 'scale' and 'constraints' parameters are documented as objects but lack field documentation. LLMs cannot determine what keys to pass (e.g., is it monthly_requests or requests_per_month?).
get_referral_code vendor parameter does not document fuzzy matching behavior or fallback. If 'Supabase Inc' is passed but the server expects 'Supabase', the call fails silently.