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

50%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/agent-tool-builder/SKILL.md

Quality

Content

47%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 body is a reasonable overview with good progressive disclosure to a single detailed reference, but it lacks an explicit sequenced workflow with validation checkpoints and its inline code example is incomplete and improperly fenced. Concrete, executable guidance is largely deferred to the reference file.

Suggestions

Add a short numbered workflow with explicit validation checkpoints (e.g., 1. define schema, 2. write comprehensive descriptions, 3. add error handling, 4. validate inputs, 5. test with the LLM) so the body is self-contained for the core procedure.

Move the inline Python example into a proper ```python fenced block and replace '# Implementation' stubs with minimal real return logic so it is copy-paste executable.

Drop the verbatim repeat of the description in the opening paragraph to avoid redundancy and save tokens.

Dimension	Reasoning	Score
Conciseness	Mostly efficient, but the opening paragraph repeats the frontmatter description verbatim and the inline Python example is sizeable, so there is notable redundancy that could be tightened; not 4 because the verbatim repeat is clear over-explanation, not 2 because the rest is reasonably lean.	3 / 5
Actionability	Provides a concrete Python example but it is wrapped in '"""' rather than a real code fence and contains '# Implementation' stubs, with the substantive guidance offloaded to the reference; fits 'some concrete guidance but incomplete / missing key details', not 4 which needs mostly executable ready code.	3 / 5
Workflow Clarity	The body presents no sequenced multi-step workflow with validation checkpoints — it only says to read the detailed guide; per the cap guidance, a tool-building skill that can involve destructive/batch operations without inline validation steps cannot exceed 3, and here even a rough ordered sequence is largely absent, placing it at 2.	2 / 5
Progressive Disclosure	The body is an overview that clearly signals a single one-level-deep reference (references/detailed-guide.md, verified to exist) with sectioned structure; not 5 because the large inline Python example and the long 'When to Use' list arguably belong partly in the reference, leaving minor organization gaps.	4 / 5
	Total	12 / 20 Passed

Description

53%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 clearly states what the skill covers but omits any 'when to use' trigger guidance, which caps completeness and weakens trigger-term quality. It is reasonably distinct from other skills but would benefit from explicit activation phrasing.

Suggestions

Add an explicit 'Use when...' clause listing concrete triggers (e.g., 'Use when the user asks to build a tool for an agent, define a function/schema, or work with MCP tools').

Include natural synonyms and file/term variations users actually say ('function calling', 'tool_use', 'input_schema', 'MCP server') to improve trigger-term coverage.

Tighten the opening framing to lead with concrete actions (designing schemas, writing descriptions, validation, error handling) rather than the abstract 'tools are how agents interact with the world' lead.

Dimension	Reasoning	Score
Specificity	Names the domain (tool design) and a couple concrete actions ('from schema to error handling') but mostly frames what tools are rather than listing concrete capabilities, fitting the 'names domain and 1-2 concrete actions' anchor; not 4 because it doesn't list several specific actions, not 2 because it does more than just name the domain.	3 / 5
Completeness	Provides a clear 'what' (tool design from schema to error handling) but no 'when' trigger clause at all; per the missing-trigger-guidance cap this lands at 3, not 4 which requires an explicit 'when'.	3 / 5
Trigger Term Quality	Contains a few relevant keywords ('tool design', 'schema', 'error handling') but misses natural phrases a user would say like 'function calling', 'build a tool', or 'MCP', and lacks synonyms; fits between anchors 2 and 3, leaning to 3 because some relevant terms are present.	3 / 5
Distinctiveness Conflict Risk	The 'agent tool design' niche is mostly distinct with only minor overlap risk against closely related coding skills; not 5 because the broad framing could brush against general development skills.	4 / 5
	Total	13 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: sickn33/agentic-awesome-skills
Path: skills/agent-tool-builder/SKILL.md
Commit: b1aebac

Reviewed: ~15 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.