debugging-antigravity

Authoritative guide for Antigravity 2.x MCP authority, duplication forensics, UI-profile isolation, and sqlite workspace recovery. Triggers: Use when the Antigravity MCP panel spins indefinitely, MCP processes appear duplicated, sqlite workspace state throws a `__store` null error, `--user-data-dir` scope is unclear, or global `~/.gemini/config/mcp_config.json` vs workspace `.agents/mcp_config.json` ownership must be established.

Quality

86%

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

72%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.

The body is an intentionally minimal progressive-disclosure pointer: maximally concise and well-structured, but it delegates all executable detail and workflow sequencing to the reference, leaving the body's own actionability and workflow clarity incomplete.

Suggestions

Add one or two concrete first-step commands inline (e.g., the `pgrep`/`ps` census or the sqlite read-only probe) so Claude can begin diagnosing before opening the reference.

Surface a brief ordered checklist mirroring the reference's five sections so the high-level workflow is visible from the SKILL.md body alone.

State the validation/verification expectation up front (e.g., 'confirm a fresh process census and server healthchecks before declaring resolution') to anchor the feedback loop.

Dimension	Reasoning	Score
Conciseness	The two-line body assumes Claude's competence and spends no tokens on background; every word orients and points to the reference, matching the lean/efficient anchor.	3 / 3
Actionability	It gives a concrete pointer ("read and follow `.agents/skills/debugging-antigravity/references/debugging-guide.md` completely") but the body itself contains no executable commands, code, or examples — the actionable substance lives entirely in the reference.	2 / 3
Workflow Clarity	The body names the triggering scenarios ("MCP ownership, process duplication, or workspace UI crashes") but provides no step sequence or validation checkpoints; the underlying task is multi-step (five diagnostic sections in the reference), so the simple-skill carve-out does not apply.	2 / 3
Progressive Disclosure	A lean overview pointing one level deep to a single, clearly signaled reference file that exists in the bundle (`references/debugging-guide.md`), with no nested reference chains — a clean disclosure structure.	3 / 3
	Total	10 / 12 Passed

Description

100%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, third-person description that pairs a precise capability statement with an explicit, scenario-rich trigger clause tailored to a niche debugging domain. It is specific, complete, and low-conflict without resorting to filler.

Dimension	Reasoning	Score
Specificity	Enumerates four concrete capability areas — "MCP authority, duplication forensics, UI-profile isolation, and sqlite workspace recovery" — rather than vague abstractions, matching the multi-action anchor.	3 / 3
Completeness	Clearly states what the skill covers (the four capability areas) and when to use it via an explicit "Triggers: Use when..." clause, satisfying both halves of the anchor.	3 / 3
Trigger Term Quality	Explicit "Triggers:" clause lists natural problem-state phrasings a user would report ("MCP panel spins indefinitely", "MCP processes appear duplicated", "__store null error", "--user-data-dir scope is unclear"), giving strong domain coverage.	3 / 3
Distinctiveness Conflict Risk	Scoped tightly to "Antigravity 2.x" MCP/version-specific failure modes with product-specific paths, making collision with unrelated skills unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: neomjs/neo
Path: .agents/skills/debugging-antigravity/SKILL.md
Commit: b4a1b58

Reviewed: 8 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.