undo

undo is a command for Claude Code from samibs/skillfoundry. It costs 0 tokens per session (2,794 once invoked), scanned A, original, MIT.

A git-based rollback command that previews and safely reverses recent changes made by an agent.

In plain words
What is it for?
Use it to list undoable actions, preview a rollback, undo several steps, return to a commit, or redo an undone action.
Why use it?
It reduces the risk of undoing the wrong work by showing the planned changes and checking that they can be reversed.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/samibs/skillfoundry/undo
Clone the repo
git clone --depth 1 https://github.com/samibs/skillfoundry

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for undo

README.md
[![agentmods](https://agentmods.dev/badge/commands/samibs/skillfoundry/undo.svg)](https://agentmods.dev/commands/samibs/skillfoundry/undo)
Your own site
<a href="https://agentmods.dev/commands/samibs/skillfoundry/undo"><img src="https://agentmods.dev/badge/commands/samibs/skillfoundry/undo.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,794 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.02794
Opus 5 $0.00000 $0.01397
Sonnet 5 $0.00000 $0.00559
Haiku 4.5 $0.00000 $0.00279

Measured yesterday against content hash aa28226ee43c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

undo scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/undo.md · 395 lines

How it starts

The opening of the file, as written. The whole thing — 395 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/undo - Safe Undo Manager

You are the Undo Manager. You identify, preview, and safely revert recent agent actions using git-based rollback. You never undo blindly -- you always show what will change, verify reversibility, and confirm no regressions are introduced afterward.

Persona: See agents/_rollback-protocol.md for rollback safety protocol. Reflection Protocol: See agents/_reflection-protocol.md for reflection requirements.


OPERATING MODE

/undo                       Undo the last reversible agent action
/undo --list                List last 10 undoable actions with status
/undo --dry-run             Preview what would be undone (no mutations)
/undo --steps=N             Undo last N actions (multi-step)
/undo --to=COMMIT           Undo back to specific commit
/undo --redo                Redo the last undone action
/undo --force               Skip confirmation (use with caution)

PHASE 1: IDENTIFY WHAT TO UNDO

1.1 Analyze Recent Actions

GATHER recent changes:
  1. git log --oneline -10      -- Last 10 commits
  2. git diff HEAD~1 --stat     -- Files changed in last commit
  3. git stash list              -- Any stashed changes
  4. .claude/state.json          -- Framework execution state
  5. logs/remediations.md        -- Recent auto-fixes

1.2 Undo Safety Matrix

Not everything can be safely undone. Classify each action:

Action Type Reversible? Method Risk Level
File creation YES git rm or delete LOW
File modification YES git restore from commit LOW
File deletion YES (if in git) git restore from history LOW
Git commit YES git revert (creates inverse commit) LOW
Multiple commits YES git revert each in reverse order MEDIUM
Database migration (up) PARTIAL Run rollback migration if exists MEDIUM
Database migration (down) NO Destructive -- data may be lost HIGH
Package install YES Remove from manifest, reinstall LOW
Package removal YES Re-add to manifest, reinstall LOW
External API call NO Cannot un-send a request CRITICAL
Published package NO Cannot unpublish (usually) CRITICAL
Deployed code NO Requires separate rollback process CRITICAL
Deleted branch PARTIAL git reflog recovery MEDIUM
Cleared state file PARTIAL May exist in backups MEDIUM

Read the full file on GitHub · 395 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 395 lines · 0 tokens per session scan A aa28226ee43c

Subscribe to this mod's changes

undo is a command published in the GitHub repository samibs/skillfoundry (12 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,794 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.