agent-tool-builder

Tools are how AI agents interact with the world. A well-designed tool is the difference between an agent that works and one that hallucinates, fails silently, or costs 10x more tokens than necessary. This skill covers tool design from schema to error handling.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

65%

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

The body is a rich, highly actionable reference with excellent code and a concrete validation checklist, but it is a verbose monolith with conceptual padding and no progressive disclosure into separate files. Actionability is its standout strength; conciseness, workflow sequencing, and file structure sit at the mid-level.

Suggestions

Trim conceptual prose Claude already knows (the MCP intro, the generic 'MCP Benefits' bullet list, and the repeated intro pitch) to tighten token efficiency.

Split large blocks (MCP implementation, the full validation checklist, parallel-execution details) into reference files and link them one level deep from a concise overview in SKILL.md.

Add at least one explicit sequenced workflow with a validation checkpoint (e.g. define schema -> validate against checklist -> test with LLM -> ship) to make the process loop clear.

Dimension	Reasoning	Score
Conciseness	The ~700-line body is mostly actionable but padded with conceptual prose Claude already knows (e.g. 'MCP ... is Anthropic's open standard for connecting AI agents to external systems. Build once, use everywhere.') and generic benefit lists, and repeats the description's pitch in the intro. It is not a 3 due to verbosity and redundant explanation, and not a 1 because substantial concrete reference content earns its place.	2 / 3
Actionability	Provides fully executable Python and TypeScript code, concrete BAD/GOOD JSON Schema examples, a copy-paste error-handling implementation, and a validation checklist with severities and exact messages. It is not a 2 because guidance is concrete and complete rather than pseudocode, and it clearly meets the 'copy-paste ready' anchor.	3 / 3
Workflow Clarity	Content is organized as patterns and a validation checklist rather than a sequenced multi-step workflow with explicit checkpoints; the error-handling pattern is a code template, not a validate->fix->retry loop. It is not a 3 because no explicit sequenced workflow with feedback loops is given, and not a 1 because structure and a severity-graded checklist are present.	2 / 3
Progressive Disclosure	No bundle files exist and the entire reference (schema design, examples, error handling, MCP, tool runners, parallel execution, validation) is a single monolithic inline file with no navigation to deeper material. It is not a 3 because content that should be split is inline with no one-level-deep references, and not a 1 because sections are clearly organized rather than a disorganized wall and references are not deeply nested.	2 / 3
	Total	9 / 12 Passed

Description

50%

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 frames the skill's value well but reads as a rationale pitch rather than a capability+trigger statement: it names the domain yet lacks concrete actions and any explicit 'Use when...' guidance. This leaves all four dimensions at the mid-level.

Suggestions

Replace the opening pitch with concrete capability verbs, e.g. 'Designs and implements AI agent tools: writes JSON Schemas, crafts LLM-facing descriptions, adds validation and error handling, and builds MCP servers.'

Add an explicit trigger clause, e.g. 'Use when the user asks to build, design, or debug a tool/function for an agent, define input_schema or tool_use/tool_result, or create an MCP server.'

Surface natural user trigger terms ('function calling', 'tool schema', 'MCP server', 'tool_use') so the description matches how users actually phrase the request.

Dimension	Reasoning	Score
Specificity	Names the domain ('tool design from schema to error handling') and its range, but uses abstract framing rather than listing multiple concrete actions like 'designs tool schemas' or 'implements error handling'. It is not a 3 because no concrete action verbs are enumerated, and not a 1 because a clear domain and scope are named.	2 / 3
Completeness	Partially answers 'what' (tool design from schema to error handling) but the 'when' is entirely absent with no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 2. It is not a 1 because a 'what' is present, and not a 3 because 'when' is missing.	2 / 3
Trigger Term Quality	Includes the natural domain term 'tools' that users would say, but surfaces no common variations or explicit trigger phrases such as 'build a tool', 'function calling', or 'MCP server'. It is not a 3 due to missing common natural terms, and not a 1 because 'tools'/'AI agents' are genuine user vocabulary rather than pure jargon.	2 / 3
Distinctiveness Conflict Risk	'Tool design'/'agent tools' is a reasonably distinct niche, but the broad framing 'tool design from schema to error handling' could overlap with general API-design or backend skills. It is not a 3 because triggers are not distinctive enough, and not a 1 because the agent-tool focus is non-generic.	2 / 3
	Total	8 / 12 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 (715 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: sickn33/antigravity-awesome-skills
Commit: 2aaaf59

Reviewed: about 8 hours 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.