Name: tools
Rating: 56.00000000000001 (1 reviews)
Author: launchdarkly

tools

Give your agents capabilities through tools (function calling). Helps you identify what your agent needs to do, create tool definitions, and attach them to config variations.

Quality

70%

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/agentcontrol/tools/SKILL.md

Quality

Content

71%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.

A highly actionable, well-sequenced operational skill: exact tool names, executable code, explicit verification, and thoughtful failure-mode warnings. Its weaknesses are redundancy (the PATCH-clobbering warning stated three times) and a monolithic body where per-provider conversion and agent-loop reference material is inlined instead of split into reference files.

Suggestions

Split the 'Per-provider schema at the call site' and 'Agent loop with tool calls' sections into a references/ file (e.g. references/provider-tool-shapes.md), keeping only the mapping table and a pointer in SKILL.md.

Deduplicate the 'pass only the tools field' warning — state it once in Step 3 and reference it briefly from the What NOT to Do list instead of restating the full rationale twice.

Fold the Edge Cases recovery actions (409, invalid schema) into their respective workflow steps so recovery is an explicit validate→fix→retry loop rather than a side table, and add a 'what to do if verification shows the tool is not attached' step.

Dimension	Reasoning	Score
Conciseness	The core MCP workflow (prerequisites, steps 1–4, edge cases) is tight and operational, but the 'don't bundle other fields into the PATCH' warning is repeated in full three times, 'never stop after listing' appears twice in Step 1 plus Core Principle 5, and ~110 lines of per-provider conversion and agent-loop material restate provider tool-schema shapes Claude largely already knows — more than the 'minor instances' of the score-4 anchor.	3 / 5
Actionability	Fully executable throughout: exact MCP tool names, copy-paste JSON request bodies, complete Python conversion snippets for OpenAI/Anthropic/Bedrock/Gemini, a runnable agent-loop skeleton, and a concrete edge-case table (409 handling, /ai-tools endpoint) covering the common cases.	5 / 5
Workflow Clarity	Steps 1–4 are clearly sequenced with an explicit verification step (two named checks via get-ai-tool and get-ai-config) and recovery actions for 409 and invalid schemas. It falls short of 5 because recovery guidance lives in a separate Edge Cases table rather than an integrated validate→fix→retry loop, and there is no guidance for what to do when verification shows the tool is not attached.	4 / 5
Progressive Disclosure	No bundle files exist (no references/, scripts/, or assets/), and the body's links point outside the bundle to ../built-in-metrics/references/*.md, which cannot be verified. Roughly half the body is provider-conversion and agent-loop material that reads as inline reference content — the score-3 hallmark of 'content that should be separate is inline' — even though sections are clearly headed and the links are well-signaled.	3 / 5
	Total	15 / 20 Passed

Description

57%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 names concrete, well-chosen actions but omits any 'when to use' trigger guidance and is written in second person, both of which the rubric penalizes. Trigger terms are decent but lack synonyms. A third-person rewrite with an explicit 'Use when' clause would lift it substantially.

Suggestions

Add an explicit trigger clause, e.g. 'Use when adding tools or function-calling capabilities to LaunchDarkly AI configs or variations'.

Rewrite in third person: 'Identifies needed agent capabilities, creates tool definitions, and attaches them to config variations' instead of 'Give your agents...' / 'Helps you identify...'.

Include natural synonyms users would say — 'function calling', 'tool schemas', 'MCP', 'attach tools to an agent config' — to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names several concrete actions ("identify what your agent needs to do, create tool definitions, and attach them to config variations") matching the score-4 anchor, but the second-person voice ("Give your agents capabilities", "Helps you identify") triggers the rubric's 1-point specificity penalty, bringing it to 3.	3 / 5
Completeness	The 'what' is clear (identify needs, create tool definitions, attach to config variations), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which the rubric explicitly caps at 3.	3 / 5
Trigger Term Quality	Good natural-term coverage — "tools", "function calling", "tool definitions", "config variations", "agents" are phrases users would say in this domain — though common synonyms (e.g., MCP, tool schemas, attach tools) are missing, so it falls short of the comprehensive-synonym bar of 5.	4 / 5
Distinctiveness Conflict Risk	"Tools", "agents", and "function calling" are broad terms that overlap with sibling agent-config skills (the body itself lists configs-create and configs-variations); only "config variations" narrows the niche, leaving moderate overlap risk.	3 / 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
relative_links	Relative link issues: 1 missing, 5 suspicious	Warning

	Total	15 / 16 Passed

Repository: launchdarkly/ai-tooling
Path: skills/agentcontrol/tools/SKILL.md
Commit: ef54971

Reviewed: 6 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.