agentic-development

Build AI agents with Pydantic AI (Python) and Claude SDK (Node.js)

Quality

65%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/agentic-development/SKILL.md

Quality

Content

70%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable with strong, explicitly-validated workflow guidance and concrete multi-framework code. Its main weaknesses are conciseness (concept re-explanation and cross-framework duplication) and progressive disclosure (a monolithic file with no external references).

Suggestions

Split per-framework detail into separate reference files (e.g., references/pydantic-ai.md, references/claude-sdk.md, references/openai-patterns.md) and keep SKILL.md as an overview pointing one level deep, which would lift progressive_disclosure.

Remove concept explanations Claude already knows (the 'Three Components' Model/Tools/Instructions diagram, basic descriptions of what an agentic loop is) to tighten conciseness.

Replace undefined helper calls in the Explore-Plan-Execute-Verify examples (llmCall, executeStep, verify) with concrete, runnable implementations or clearly mark them as illustrative scaffolding.

Dimension	Reasoning	Score
Conciseness	The body is roughly 850 lines with mostly code, which is efficient in places, but it re-explains basic concepts Claude already knows (e.g., the 'Three Components' ASCII diagram of Model/Tools/Instructions) and duplicates patterns across four frameworks, so it is mostly efficient but could be tightened; not a 4 due to the conceptual padding and cross-framework redundancy.	3 / 5
Actionability	Provides numerous concrete, largely executable Python and TypeScript examples (Pydantic AI agents, Claude SDK agentic loop, tool definitions); not a 5 because several workflow examples call undefined helpers (llmCall, executeTool, executeStep, verify, selfCorrect) that read as illustrative pseudocode with minor gaps.	4 / 5
Workflow Clarity	The Explore-Plan-Execute-Verify workflow is explicitly sequenced with a dedicated Verify phase, per-step verification ('Verify before continuing'), self-correction, escalation (handleFailure), plus a final checklist — matching the anchor for clear sequence with explicit validation steps and feedback loops.	5 / 5
Progressive Disclosure	Content is well-sectioned with clear headers, but the file is a large monolith (~850 lines) with no bundle files and no references to separate files; per-framework deep dives and API/reference material that clearly belongs in separate files are all inlined, which fits the 'some structure but content that should be separate is inline' anchor.	3 / 5
	Total	15 / 20 Passed

Description

61%Weight 40%Scale 1-5

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 specific and names concrete frameworks, but the evaluated description field lacks an explicit 'Use when' trigger clause, which caps completeness. Distinctiveness is reasonably strong thanks to the named frameworks.

Suggestions

Move the trigger guidance into the description field itself with a 'Use when...' clause (e.g., '...Use when building AI agents, tool-using LLM systems, or agentic workflows in Python or Node.js') so completeness is not capped at 3.

Add 1-2 more concrete actions beyond 'Build' (e.g., 'design, build, and test AI agents') to lift specificity toward the 4-5 anchors.

Include additional natural trigger synonyms (e.g., 'agent SDK', 'LLM tools') to push trigger term coverage from good to comprehensive.

Dimension	Reasoning	Score
Specificity	Names the domain ('AI agents') and one concrete action ('Build') plus two specific frameworks/languages (Pydantic AI/Python, Claude SDK/Node.js), but does not enumerate multiple distinct actions, so it sits at the '1-2 concrete actions, not comprehensive' anchor rather than the 4 which requires several specific actions.	3 / 5
Completeness	The description field gives a clear 'what' ('Build AI agents with Pydantic AI (Python) and Claude SDK (Node.js)') but contains no 'Use when...' clause, so per the guideline a missing explicit trigger clause caps completeness at 3 even though a separate when-to-use frontmatter field exists.	3 / 5
Trigger Term Quality	'building AI agents', 'tool-using LLM systems', 'agentic workflows' are natural phrases users would say, giving good keyword coverage with synonyms; not a 5 because it lacks the fuller set of natural variations and explicit triggers.	4 / 5
Distinctiveness Conflict Risk	Naming two specific frameworks (Pydantic AI, Claude SDK) tied to languages gives it a clear, mostly-distinct niche with only minor overlap risk against generic agent skills; not a 5 because 'AI agents' is still a broad trigger that could overlap with other agent-building guidance.	4 / 5
	Total	14 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 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	14 / 16 Passed

Repository: alinaqi/claude-bootstrap
Path: skills/agentic-development/SKILL.md
Commit: abc600e

Reviewed: 12 days ago

Table of Contents

Discovery Implementation Validation

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.