agentic-development
Build AI agents with Pydantic AI (Python) and Claude SDK (Node.js)
33
30%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/agentic-development/SKILL.mdQuality
Discovery
32%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description is too terse—it names the domain and two frameworks but lacks concrete actions, explicit trigger guidance, and natural user keywords. It would benefit significantly from listing specific capabilities and adding a 'Use when...' clause to help Claude distinguish this skill from other coding or AI-related skills.
Suggestions
Add a 'Use when...' clause with trigger terms like 'Use when the user wants to build an AI agent, create tool-calling workflows, or use Pydantic AI or the Anthropic Claude SDK'.
List specific concrete actions such as 'define agent tools, configure agent loops, handle structured outputs, manage conversation state, implement streaming responses'.
Include natural keyword variations users might say: 'agentic workflow', 'tool use', 'function calling', 'pydantic-ai', 'anthropic SDK', 'agent framework'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (AI agents) and specifies two frameworks (Pydantic AI, Claude SDK) with their languages, but doesn't list concrete actions like 'create tool definitions', 'configure agent loops', 'handle streaming responses', etc. | 2 / 3 |
Completeness | Provides a brief 'what' (build AI agents with specific frameworks) but completely lacks a 'Use when...' clause or any explicit trigger guidance for when Claude should select this skill. Per rubric guidelines, missing 'Use when' caps completeness at 2, and the 'what' is also thin, so this scores 1. | 1 / 3 |
Trigger Term Quality | Includes some relevant keywords like 'AI agents', 'Pydantic AI', 'Claude SDK', 'Python', 'Node.js', but misses common user terms like 'agentic', 'tool use', 'function calling', 'agent framework', 'pydantic-ai', or 'anthropic SDK'. | 2 / 3 |
Distinctiveness Conflict Risk | Mentioning specific frameworks (Pydantic AI, Claude SDK) provides some distinctiveness, but 'Build AI agents' is broad enough to overlap with other agent-building or SDK-related skills. The dual-framework scope also increases potential conflict with Python-specific or Node.js-specific skills. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
27%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is far too long and broad for its stated purpose of building agents with Pydantic AI and Claude SDK. It tries to be a comprehensive agent development encyclopedia covering OpenAI, Gemini, Claude, and general patterns, resulting in massive token waste. The most valuable parts—the concrete Pydantic AI and Claude SDK code examples—are buried in a sea of generic architecture advice, undefined abstractions, and multi-framework coverage that dilutes focus.
Suggestions
Reduce scope to match the description: focus only on Pydantic AI (Python) and Claude SDK (Node.js), removing OpenAI, Gemini, and generic agent architecture content.
Split into multiple files: create separate reference files for Pydantic AI patterns, Claude SDK patterns, tool design, and testing, with SKILL.md as a concise overview with navigation links.
Replace pseudocode abstractions (executeStep, selfCorrect, llmCall, createAgent) with actual executable code using the two target frameworks.
Remove content Claude already knows: the ASCII architecture diagram, explanations of what agents/tools/memory are, basic project structure templates, and generic software engineering advice like 'test edge cases'.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~600+ lines. Explains concepts Claude already knows (what agents are, what tools are, basic architecture diagrams). Includes extensive coverage of OpenAI, Gemini, and general agent patterns that go far beyond the stated scope of 'Pydantic AI (Python) and Claude SDK (Node.js)'. Much content is generic software engineering advice (project structures, testing patterns, memory interfaces) that doesn't earn its token cost. | 1 / 3 |
Actionability | Contains many code examples that appear executable (Pydantic AI agent setup, Claude SDK agentic loop, tool definitions), which is good. However, much of the code is illustrative/pseudocode-like (e.g., `executeStep`, `selfCorrect`, `llmCall`, `createAgent` are undefined functions). The Pydantic AI and Claude SDK examples at the top are the most concrete and copy-paste ready, but the workflow and guardrail sections rely on hypothetical interfaces. | 2 / 3 |
Workflow Clarity | The Explore-Plan-Execute-Verify workflow is clearly sequenced with verification steps, which is good. However, the verification code relies on undefined abstractions (VerificationSchema, runCommand, etc.), making it hard to actually follow. The skill lacks concrete validation checkpoints for the actual development process—it describes what an agent should do but not how to validate the agent itself during development. | 2 / 3 |
Progressive Disclosure | Monolithic wall of text with no references to external files despite the massive length. Everything is inlined—project structures, multiple framework patterns, anti-patterns, model selection guides, testing patterns, and multiple language examples. This content desperately needs to be split into separate files (e.g., pydantic-ai.md, claude-sdk.md, patterns.md, testing.md) with the SKILL.md serving as a concise overview with navigation links. | 1 / 3 |
Total | 6 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (856 lines); consider splitting into references/ and linking | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Repository
- alinaqi/claude-bootstrap
- Commit
65efb33
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.