Back to search
ai-product-skills
MarketplaceWorks with Claude CodeGitHub
Agent skills for building AI products: Gemini prompts, agent behaviour changes with evals, FastAPI/ADK/LangGraph backends, AI code review. Each skill tested with vs without.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/sufianwajid--ai-product-skills)HTML
<a href="/listings/sufianwajid--ai-product-skills"><img src="/api/listings/sufianwajid--ai-product-skills/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Claude Code (plugin)
/plugin marketplace add SufianWajid/ai-product-skills /plugin install ai-product-skills@ai-product-skills
Claude Code or Cursor (copy)
git clone https://github.com/SufianWajid/ai-product-skills # all projects (Claude Code): cp -r ai-product-skills/skills/* ~/.claude/skills/ # one project: cp -r ai-product-skills/skills/<skill-name> .claude/skills/ # Claude Code cp -r ai-product-skills/skills/<skill-name> .cursor/skills/ # Cursor
Skills load automatically when your request matches their description. You can also ask for one by name, for example "use ai-code-review on this PR".
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
.claude-plugin/marketplace.json · Manifest · 26 lines
- {
- "name": "ai-product-skills",
- "owner": { "name": "Mian Sufian Wajid" },
- "metadata": {
- "description": "Agent skills for building AI products: Gemini prompts, agent behaviour changes and evals, FastAPI/ADK/LangGraph backends, AI code review.",
- "version": "0.1.0"
- },
- "plugins": [
- {
- "name": "ai-product-skills",
- "description": "Agent skills for building and shipping LLM products in production.",
- "source": "./",
- "strict": false,
- "skills": [
- "./skills/gemini-prompt-writing",
- "./skills/agent-behaviour-change",
- "./skills/ai-code-review",
- "./skills/fastapi-gemini-agent",
- "./skills/adk-openapi-tools",
- "./skills/langgraph-fastapi-architecture",
- "./skills/express-typescript-modular",
- "./skills/firebase-project-infra"
- ]
- }
- ]
- }
README.md · README · 135 lines
- # AI product skills
- Agent skills for building and shipping LLM products in production, for
- [Claude Code](https://docs.claude.com/en/docs/claude-code/skills), Cursor, and any tool that
- reads the open `SKILL.md` format.
- These came out of shipping real AI products: a voice companion for children with a
- safety classifier and an evaluation bench, Gemini agent backends on FastAPI, a Google ADK
- assistant with read-only data tools, and LangGraph multi-agent services. Each skill is the
- working method from those projects, generalised so it fits any codebase.
- ## What difference does it make?
- Same request to Claude, with and without the `fastapi-gemini-agent` skill:
- > *"I need a small FastAPI service with one endpoint, POST /agent/ask, where a customer asks
- > about their order in plain English and a Gemini agent answers. It has one tool,
- > get_order_status(order_id)… make it production-ready, with tests."*
- **Without the skill**, Claude hands the tool loop to the SDK, so nothing checks which tool the
- model calls:
- ```python
- config = types.GenerateContentConfig(
- tools=[get_order_status],
- automatic_function_calling=types.AutomaticFunctionCallingConfig(maximum_remote_calls=3),
- )
- ```
- **With the skill**, it writes the loop itself, capped and allow-listed:
- ```python
- automatic_function_calling=types.AutomaticFunctionCallingConfig(disable=True)
- ...
- for _ in range(self._s.max_turns): # bounded: never loop forever
- ...
- if unknown := [c.name for c in calls if c.name not in TOOLS]:
- raise UnknownTool(", ".join(map(str, unknown)))
- ```
- | | Without skill | With skill |
- |---|---|---|
- | Unknown tool names | not checked | rejected before anything runs |
- | Empty model answer | returned as a 200 success | treated as an error (502) |
- | Error responses | FastAPI defaults, mixed shapes | one consistent shape with a request id |
- | Structure | one `main.py` | thin route, one Gemini service, settings, prompt template |
- | Tests | 11, Gemini faked | 27, including unknown tool, timeout and turn limit (run and passing) |
- | **Score** | **2/8 checks** | **8/8 checks** |
- ## Skills
- | Skill | Use it when you're… |
- |---|---|
- | [gemini-prompt-writing](skills/gemini-prompt-writing/SKILL.md) | writing or fixing a production prompt for Gemini: structure, structured output, token budget, no-example rules |
- | [agent-behaviour-change](skills/agent-behaviour-change/SKILL.md) | changing what a live agent says, and proving the change on an eval bench before release |
- | [ai-code-review](skills/ai-code-review/SKILL.md) | reviewing an LLM app PR for bugs, dead code, over-engineering and AI bloat in prompts and docs |
- | [fastapi-gemini-agent](skills/fastapi-gemini-agent/SKILL.md) | building a FastAPI agent service on the `google-genai` SDK with a bounded tool loop |
- | [adk-openapi-tools](skills/adk-openapi-tools/SKILL.md) | giving a Google ADK agent safe, read-only, user-scoped tools from FastAPI routes |
- | [langgraph-fastapi-architecture](skills/langgraph-fastapi-architecture/SKILL.md) | keeping a FastAPI + LangGraph multi-agent backend inside clean layer boundaries |
- | [express-typescript-modular](skills/express-typescript-modular/SKILL.md) | structuring an Express 5 + TypeScript API as feature modules |
- | [firebase-project-infra](skills/firebase-project-infra/SKILL.md) | setting up, securing and deploying Firebase with per-environment projects |
- ### The ideas behind them
- - **The model judges, code routes.** Behaviour comes from prompts and the context they get,
- never from regexes that inspect or rewrite replies.
- - **No examples in model-facing text.** Models copy examples. State the rule; put examples
- in the eval set.
- - **Every bug becomes a release-gate case** that fails before the fix and passes after.
- - **Prompts have a token budget,** enforced by a test, so they don't grow with every fix.
- ## Tested
- Every skill has a realistic test task in its `evals/` folder, with the input files and the
- pass/fail checks. Each task was run by two separate Claude agents, one with the skill and one
- without, and an independent grader scored both answers against the same checks.
- | Skill | With skill | Without |
- |---|---|---|
- | fastapi-gemini-agent | 8/8 | 2/8 |
- | langgraph-fastapi-architecture | 6/6 | 2/6 |
- | firebase-project-infra | 9/9 | 6/9 |
- | agent-behaviour-change | 8/8 | 5/8 |
- | adk-openapi-tools | 8/8 | 6/8 |
- | express-typescript-modular | 8/8 | 6/8 |
- | ai-code-review | 9/10 | 8/10 |
- | gemini-prompt-writing | 7/7 | 3/7 |
- Overall: **63 of 64 checks passed with the skills (98%), 38 of 64 without (59%)** (one task per
- skill, November 2026). The gemini-prompt-writing score is from the current version, run twice; an earlier
- version scored 3/7, and the evaluation is what showed it needed a cutting pass and a word budget.
- These are early results from one task per skill, not a benchmark. To rerun one (these are real,
- billed model runs):
- ```bash
- cd skills/<skill-name> && claude plugin eval . --allow-tools Write,Edit,Bash --scaffold
- ```
- ## Install
- ### Claude Code (plugin)
- ```text
- /plugin marketplace add SufianWajid/ai-product-skills
- /plugin install ai-product-skills@ai-product-skills
- ```
- ### Claude Code or Cursor (copy)
- ```bash
- git clone https://github.com/SufianWajid/ai-product-skills
- # all projects (Claude Code):
- cp -r ai-product-skills/skills/* ~/.claude/skills/
- # one project:
- cp -r ai-product-skills/skills/<skill-name> .claude/skills/ # Claude Code
- cp -r ai-product-skills/skills/<skill-name> .cursor/skills/ # Cursor
- ```
- Skills load automatically when your request matches their description. You can also ask for
- one by name, for example "use ai-code-review on this PR".
- ## Contributing
- Issues and pull requests are welcome, especially corrections from real use ("this rule failed
- for me when…").
- - One skill per folder under `skills/`, with a `SKILL.md` whose `name` matches the folder.
- - Keep `SKILL.md` short; put long material in `references/`.
- - Add new skills to `.claude-plugin/marketplace.json`.
- - Run `python3 scripts/validate.py` before opening a PR (CI runs it too).
- ## License
- [MIT](LICENSE) © Mian Sufian Wajid
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.