Safely undo a phase - roll back its commits from the SUMMARY manifest, revert (or --no-commit to re-do), and reset the phase status
Scanned 9/3/2026
Install to Claude Code
npx -y skills add crenshawdev/cadence --skill cad-undo --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Cad Undo?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/crenshawdev-cad-undo)More formats (shields.io, HTML) on the badges page.
---
name: cad-undo
description: "Safely undo a phase - roll back its commits from the SUMMARY manifest, revert (or --no-commit to re-do), and reset the phase status"
argument-hint: "<phase N> [--no-commit]"
allowed-tools:
- Read
- Edit
- Bash
- Grep
- Glob
- AskUserQuestion
---
<objective>
Undo a phase's work by reverting its commits, using the phase SUMMARY.md as the
manifest of exactly which hashes to touch (cad-execute records commits-per-task
with hashes there). Safe by construction: never runs on a dirty tree, never
guesses. No heuristic dependency-check - instead cad-undo states plainly
whether later phases exist and lets the user decide.
</objective>
<execution_context>
@${CLAUDE_PLUGIN_ROOT}/cadence-core/workflows/undo.md
@${CLAUDE_PLUGIN_ROOT}/cadence-core/references/git-guard.md
</execution_context>
<process>
Run the undo workflow end-to-end. Refuse on a dirty tree. Show the exact hashes
and get confirmation before reverting. Reset the phase's status as part of the
rollback (this is the one legitimate exception to cad-verify owning status).
</process>
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...