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.
npx agentmods add skills/d3x293/code-crew/task-rollbacknpx skills add d3x293/code-crew --skill task-rollbackgit clone --depth 1 https://github.com/d3x293/code-crewWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00041 | $0.00749 |
| Opus 5 | $0.00020 | $0.00375 |
| Sonnet 5 | $0.00008 | $0.00150 |
| Haiku 4.5 | $0.00004 | $0.00075 |
Grade A, and why
task-rollback 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.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task Rollback - Undo Last Task
Reverts file changes made by the most recent /crew task execution using git.
When to Activate
- User runs
/crew rollback
Prerequisites
.claude/crew-history.jsonmust exist with at least one entry- Project must be a git repository
- The
gitRefBeforefield should be present in the last history entry (captured by execution-orchestrator before task execution)
If history is missing: "No task history found. Nothing to roll back." If not a git repo: "Not a git repository. Rollback requires git."
Rollback Process
Step 1: Read Last Task
Read .claude/crew-history.json and get the last entry:
- Extract
filesModified— the files that were changed - Extract
gitRefBefore— the git commit hash before the task started - Extract
task— the original task description - Extract
timestamp— when the task ran
Step 2: Show Preview
Display what will be rolled back:
Rollback: "{task}" ({timestamp})
Files: {file1}, {file2}, {file3}
Revert to: {gitRefBefore}
Step 3: Ask for Confirmation
This is mandatory — never skip confirmation.
Ask: "Proceed with rollback? (y/n)"
If user says no: "Cancelled."
Step 4: Execute Rollback
Two strategies based on available data:
Strategy A: gitRefBefore is available (preferred)
git checkout {gitRefBefore} -- {file1} {file2} {file3}
This reverts only the specific files to their pre-task state without affecting other changes.
Strategy B: gitRefBefore is not available (fallback)
git diff HEAD -- {file1} {file2} {file3} # Show what will change
git checkout HEAD~1 -- {file1} {file2} {file3} # Revert to previous commit
Warning: This is less precise — only works if the task's changes are in the most recent commit.
If files have been modified since the task (additional changes on top):
- Warn the user: "Warning: {file} has been modified since the task. Rolling back will also undo those changes."
- Ask for confirmation again
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.
- yesterday First seen · 102 lines · 41 tokens per session scan A e5521217cf88
task-rollback is a skill published in the GitHub repository d3x293/code-crew (13 stars, last pushed 4mo ago), licensed MIT. It adds 41 tokens to every session and 749 once invoked, about $0.0002 per session on Opus 5. 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 2026-08-30.
Other skills, from other repositories
agile-product-owner
../../../product-team/agile-product-owner/skills/agile-product-owner/SKILL.md.
apple-hig-expert
../../../product-team/apple-hig-expert/skills/apple-hig-expert/SKILL.md.
autoresearch-agent
../../../engineering/autoresearch-agent/skills/autoresearch-agent/SKILL.md.
board-prep
../../../c-level-advisor/executive-mentor/skills/board-prep/SKILL.md.
browserstack
../../../engineering-team/playwright-pro/skills/browserstack/SKILL.md.
business-investment-advisor
../../../finance/business-investment-advisor/skills/business-investment-advisor/SKILL.md.