bat-adhoc

Run bot acceptance tests to validate MCP tools work correctly from a real AI agent's perspective. Use when testing PRs, detecting regressions, or verifying tool changes end-to-end with Claude/Gemini CLIs.

Quality

87%

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

Quality

Content

85%Weight 40%Scale 1-3

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

A highly actionable, clearly sequenced skill body with executable examples, an explicit pass/fail validation checkpoint, and a well-organized structure that defers complete reference to README.md. The only soft spot is conciseness, where some advisory prose could be trimmed.

Suggestions

Tighten the "Cost Awareness" section: replace the advisory prose with a concise bullet such as "Cost: one API call per agent per phase — combine related checks in one test_prompt and start with one agent."

Trim the standalone "Robustness tip" sentence or fold it into the Regression Comparison metrics as a one-line note, since it restates guidance already implied by the variation-testing concept.

Consider moving the full "Output Structure" JSON block to tests/uat/README.md and keeping only the `all_passed` / `results_file` fields inline, reducing the SKILL.md token footprint while preserving actionability.

Dimension	Reasoning	Score
Conciseness	The body is mostly operational and avoids explaining concepts Claude already knows, but advisory prose such as the "Cost Awareness" section and the "Robustness tip" could be tightened. It is not padded with redundant concept explanations (rule out 1) but does not have every token earning its place (rule out 3).	2 / 3
Actionability	Concrete executable commands ("python tests/uat/run_uat.py", "git fetch origin master && git checkout master && git pull") and a complete heredoc scenario example make the guidance copy-paste ready, matching the score-3 anchor. It is not pseudocode or vague (rule out 1/2).	3 / 3
Workflow Clarity	The 6-step Workflow has an explicit validation checkpoint ("Check `all_passed` per agent. If true, you're done.") and a failure feedback loop (dig deeper on failure → regression check with `--branch master`), matching the score-3 anchor. Validation is present, so the destructive/batch cap does not apply (rule out 2).	3 / 3
Progressive Disclosure	The body is organized into clear sections and defers full detail to a well-signaled one-level reference ("For complete CLI reference and output format, see `tests/uat/README.md`"). No bundle files exist to verify, but structure and navigation are clear, matching the score-3 anchor (rule out 1/2).	3 / 3
	Total	11 / 12 Passed

Description

90%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, well-formed description with an explicit "Use when" trigger clause, natural trigger terms, and a clearly distinct niche. Its only weakness is specificity: it describes one core action rather than enumerating several concrete capabilities.

Dimension	Reasoning	Score
Specificity	The description names the domain and a concrete action ("Run bot acceptance tests to validate MCP tools") but presents a single core operation rather than a list of multiple distinct concrete actions like the score-3 anchor (extract/fill/merge). It is not vague (rule out 1), yet it is not comprehensive across several specific actions (rule out 3).	2 / 3
Completeness	It explicitly answers "what" ("Run bot acceptance tests to validate MCP tools...") and "when" via an explicit "Use when testing PRs, detecting regressions, or verifying tool changes..." trigger clause, matching the score-3 anchor. It is not missing the when (rule out 2).	3 / 3
Trigger Term Quality	"testing PRs, detecting regressions, verifying tool changes end-to-end with Claude/Gemini CLIs" gives good coverage of natural phrases a user would say, exceeding the score-2 single-keyword anchor ("Works with PDF files"). The terms are natural and varied, matching the score-3 anchor.	3 / 3
Distinctiveness Conflict Risk	The niche is specific and distinct — acceptance testing MCP tools from a real AI agent's perspective via Claude/Gemini CLIs — making it unlikely to trigger for the wrong skill, matching the score-3 anchor. It is not generic enough to broadly conflict (rule out 1/2).	3 / 3
	Total	11 / 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
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: homeassistant-ai/ha-mcp
Path: .claude/skills/bat-adhoc/SKILL.md
Commit: 01964cc

Reviewed: 11 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.