CtrlK
BlogDocsLog inGet started
Tessl Logo

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 necessar...

Install with Tessl CLI

npx tessl i github:sickn33/antigravity-awesome-skills --skill agent-tool-builder
What are skills?

42

1.05x

Quality

11%

Does it follow best practices?

Impact

100%

1.05x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/agent-tool-builder/SKILL.md
SKILL.md
Review
Evals

Evaluation results

100%

Shopping Assistant Tool Definitions

Schema quality & documentation

Criteria
Without context
With context

Specific parameter descriptions

100%

100%

No generic catch-all descriptions

100%

100%

Distinguishable tool descriptions

100%

100%

Parameter types declared

100%

100%

At least one example in descriptions

100%

100%

Required vs optional parameters

100%

100%

No over-parameterisation

100%

100%

Focused tool count

100%

100%

README present

100%

100%

Without context: $0.2056 · 52s · 8 turns · 9 in / 3,173 out tokens

With context: $0.4010 · 1m 49s · 18 turns · 67 in / 5,925 out tokens

100%

15%

File Management Agent Tools

Error handling & recovery

Criteria
Without context
With context

Error message present

100%

100%

No silent catch blocks

86%

100%

Error includes recovery hint

46%

100%

Distinguishes error types

100%

100%

Error structure consistent

100%

100%

Errors documented in schema

50%

100%

design_notes.md present

100%

100%

Successful path returns useful data

100%

100%

Without context: $0.3328 · 1m 19s · 14 turns · 14 in / 4,967 out tokens

With context: $0.5829 · 2m 25s · 25 turns · 73 in / 8,654 out tokens

100%

GitHub Repository Explorer Agent

Input examples & tool count

Criteria
Without context
With context

Examples in descriptions

100%

100%

Example format is specific

100%

100%

Focused tool count

100%

100%

No duplicated responsibilities

100%

100%

Parameter descriptions not generic

100%

100%

Flexible parameters documented

100%

100%

tool_overview.md present

100%

100%

Required fields declared

100%

100%

Without context: $0.3034 · 1m 27s · 11 turns · 60 in / 5,367 out tokens

With context: $0.3826 · 1m 48s · 16 turns · 16 in / 5,950 out tokens

Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.