conductor-revert
Git-aware undo by logical work unit (track, phase, or task)
69
55%
Does it follow best practices?
Impact
96%
1.60xAverage score across 3 eval scenarios
Passed
No findings from the security scan
Fix and improve this skill with Tessl
tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/conductor-revert/SKILL.mdThe canonical home for this skill is conductor-revert in sickn33/agentic-awesome-skills
Evaluation results
90%
39%Authentication Refactor Phase Rollback
Phase commit discovery and conflict handling
Phase argument format
0%
100%
Phase commit discovery command
0%
100%
Plan.md update commits included
0%
50%
Execution plan before reverting
70%
100%
YES confirmation prompt
20%
100%
Reverse chronological execution
100%
100%
HALT on conflict
100%
100%
Conflict options displayed
20%
100%
plan.md phase tasks reset
100%
50%
tracks.md status updated
100%
100%
100%
37%Dashboard Widget Rollback
Task-level revert with pre-flight checks and plan updates
Conductor pre-flight
0%
100%
Git status pre-flight
100%
100%
Uncommitted changes warning
30%
100%
Execution plan displayed
100%
100%
YES confirmation required
0%
100%
git revert --no-edit used
100%
100%
Reverse chronological order
100%
100%
Task commit discovery command
0%
100%
plan.md task marker updated
100%
100%
metadata.json updated
100%
100%
100%
33%Payment Gateway Integration Abandonment
Full track revert with safety rules and edge cases
Full track discovery command
100%
100%
Track creation commits command
0%
100%
Remote-pushed commits warning
100%
100%
Already-reverted commits handling
100%
100%
Execution plan before reverting
70%
100%
YES confirmation required
0%
100%
No git reset --hard
100%
100%
No git push --force
100%
100%
tracks.md status set to [ ]
0%
100%
Track directory deletion offered
100%
100%
- Commit
7520423
- 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.