conductor-revert

Git-aware undo by logical work unit (track, phase, or task)

Quality

63%

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/conductor-revert/SKILL.md

Quality

Content

76%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 content is highly actionable and the multi-step destructive workflow is sequenced with strong validation checkpoints and feedback loops, satisfying the destructive-operation bar. Its weakness is conciseness (heavy boxed-banner padding) and progressive disclosure (everything inlined, no supporting files split out).

Suggestions

Trim the large ASCII banners (execution plan, merge conflict, completion) to essential fields, or move the full template text into a referenced file, to reduce token padding.

Extract the verbose menu/edge-case dialog templates into a references/ file (e.g., DIALOGS.md) and link to it one level deep to improve progressive disclosure.

Consider collapsing the repeated safety-rule restatements (confirmation and halt rules appear in multiple sections) into the single Safety Rules section.

Dimension	Reasoning	Score
Conciseness	The body is mostly procedural and earns its tokens with real git commands and concrete menus, but it leans verbose: large boxed ASCII banners (execution plan, conflict, completion) and repeated warning text add padding that could be tightened without losing clarity.	3 / 5
Actionability	Concrete, executable commands are given throughout (git log --grep filters, git revert --no-edit per SHA, plan.md marker edits, metadata.json decrements) with realistic examples covering task, phase, and full-track cases, making it copy-paste ready.	5 / 5
Workflow Clarity	Pre-flight checks, target selection, commit discovery, an explicit execution-plan display with mandatory 'YES' confirmation, ordered revert execution, conflict halting, plan.md updates, and a verification section form a fully sequenced workflow with explicit validation checkpoints and feedback loops for this destructive/batch operation.	5 / 5
Progressive Disclosure	There are no bundle files (references/, scripts/, assets/ are absent) and the body is a single self-contained document with clear section headers, but detailed reference material (the full ASCII menus and edge-case dialogs) is inlined rather than split out, leaving structure at the midpoint.	3 / 5
	Total	16 / 20 Passed

Description

50%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 is concise and names a clear, distinct niche with concrete work-unit granularity, but it lacks any explicit 'Use when' trigger guidance and omits the natural term 'revert'. These gaps keep specificity, trigger quality, and completeness at the midpoint.

Suggestions

Add an explicit 'Use when...' clause naming the trigger phrases users would say (e.g., 'Use when reverting a Conductor track, phase, or task, or when the user asks to undo a logical unit of work').

Include the synonym 'revert' alongside 'undo' so the description matches the natural language users and Claude would actually invoke.

Optionally name the concrete actions the skill performs (find commits, revert them in reverse order, update plan.md) to lift specificity above 3.

Dimension	Reasoning	Score
Specificity	"undo by logical work unit (track, phase, or task)" names the domain and concrete granularities, but only describes a single action (undo) rather than multiple concrete operations, matching the score-3 anchor of domain plus limited concrete actions.	3 / 5
Completeness	The "what" is clear (git-aware undo by track/phase/task), but there is no "Use when..." or equivalent trigger clause; per the judging guidelines a missing explicit trigger caps completeness at 3.	3 / 5
Trigger Term Quality	It includes the natural term "undo" and "git-aware", plus the unit names, but omits the common synonym "revert" that the skill actually performs, so keyword coverage is partial rather than comprehensive.	3 / 5
Distinctiveness Conflict Risk	"Git-aware undo by logical work unit (track, phase, or task)" carves a distinct niche tied to the Conductor track model with low overlap risk, though it could still brush against generic git-revert guidance, placing it just above the midpoint anchor.	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
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: rmyndharis/antigravity-skills
Path: skills/conductor-revert/SKILL.md
Commit: 3eff4af

Reviewed: 1 day 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.