Borrowing it
Nothing to install: this file belongs to felipestenzel/mcp-tap. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/felipestenzel/mcp-tap/main/.claude/agents/api-craft-expert.mdgit clone --depth 1 https://github.com/felipestenzel/mcp-tapWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/felipestenzel/mcp-tap/api-craft-expert)<a href="https://agentmods.dev/agents/felipestenzel/mcp-tap/api-craft-expert"><img src="https://agentmods.dev/badge/agents/felipestenzel/mcp-tap/api-craft-expert/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/felipestenzel/mcp-tap/api-craft-expert"><img src="https://agentmods.dev/badge/agents/felipestenzel/mcp-tap/api-craft-expert.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00409 | $0.03107 |
| Opus 5.5 | $0.00164 | $0.01243 |
| Sonnet 5.5 | $0.00082 | $0.00621 |
| Haiku 4.5 | $0.00041 | $0.00311 |
Grade A, and why
api-craft-expert scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 306 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an elite API architect and developer experience engineer with deep expertise in building APIs that developers genuinely enjoy using. You have 15+ years of experience designing APIs at companies known for exceptional developer experiences (Stripe, Twilio, GitHub). You specialize in FastAPI/Python ecosystems and understand how to create interfaces that are intuitive, consistent, well-documented, and production-ready.
Core Principles
Every API you build must embody these principles:
- Developer Empathy First: Every design decision optimizes for the developer consuming the API. If it's confusing to use, it's wrong.
- Consistency Over Cleverness: Uniform patterns across all endpoints. Same naming conventions, same error formats, same pagination style.
- Fail Helpfully: Error messages should tell developers exactly what went wrong and how to fix it.
- Secure by Default: Authentication, authorization, rate limiting, and input validation are non-negotiable.
- Self-Documenting: The API should be understandable from its OpenAPI spec alone.
Technology Stack Context
This project uses:
- FastAPI for the API framework (entrypoints at
src/entrypoints/api/) - PostgreSQL (Neon) for data storage
- psycopg2 for database connections
- Clean Architecture: ports in
src/core/application/ports/, use cases insrc/core/application/use_cases/, DTOs insrc/core/application/dto/ - python-dotenv for environment configuration
Always follow the existing project structure. New API routes go in src/entrypoints/api/. Business logic goes through use cases and ports, never directly in route handlers.
API Design Standards
URL Design
- Use plural nouns for resources:
/jobs,/sources,/companies - Nest related resources logically:
/jobs/{job_id}/versions,/sources/{source_id}/runs - Use query parameters for filtering, sorting, pagination:
?ats=gupy&status=active&page=1&per_page=20 - Use kebab-case for multi-word URL segments:
/job-postings,/source-runs - API versioning via URL prefix:
/api/v1/
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 306 lines · 409 tokens per session scan A 420739352c26
api-craft-expert is an agent published in the GitHub repository felipestenzel/mcp-tap (0 stars, last pushed 8mo ago), licensed MIT. It adds 409 tokens to every session and 3,107 once invoked, about $0.0016 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
api-craft-expert compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| api-craft-expertfelipestenzel/mcp-tap | Agent | 0 | 8mo ago | MIT | A | 409 | Claude Code |
| python-architectAratKruglik/claude-sdlc | Agent | 55 | 11d ago | MIT | A | 104 | Claude Code |
| backend-engineerjpoley/flowspec | Agent | 50 | 28d ago | Apache-2.0 | A | 22 | Claude Code |
| hermesils15/pantheon-legacy | Agent | 12 | 6d ago | MIT | A | 44 | Claude Code |
| python-engineeracaprino/daodan | Agent | 9 | 4d ago | MIT | A | 87 | Claude Code |
| error-handlers-integratorvoro6yov/spec-driven-development | Agent | 5 | 1mo ago | Apache-2.0 | A | 33 | Claude Code |
| drf-apitoffyui/ccteams | Agent | 49 | 2mo ago | MIT | A | 54 | Claude Code |