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 skills add wan-huiyan/agent-traffic-control --skill dispatched-bash-agent-git-checkout-clobbers-uncommitted-editgit clone --depth 1 https://github.com/wan-huiyan/agent-traffic-controlWrote 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.
[](https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/dispatched-bash-agent-git-checkout-clobbers-uncommitted-edit)<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/dispatched-bash-agent-git-checkout-clobbers-uncommitted-edit"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/dispatched-bash-agent-git-checkout-clobbers-uncommitted-edit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/dispatched-bash-agent-git-checkout-clobbers-uncommitted-edit"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/dispatched-bash-agent-git-checkout-clobbers-uncommitted-edit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00409 | $0.02330 |
| Opus 5 | $0.00204 | $0.01165 |
| Sonnet 5 | $0.00082 | $0.00466 |
| Haiku 4.5 | $0.00041 | $0.00233 |
Grade A, and why
dispatched-bash-agent-git-checkout-clobbers-uncommitted-edit 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 11d ago.
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 — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.
A dispatched Bash agent's git checkout reverts your uncommitted worktree edit
Problem
You dispatch one or more Bash-capable subagents (a Workflow fan-out, Explore,
general-purpose, code-reviewer, etc.) to review/verify work in the same git
worktree where you have uncommitted edits. An agent, investigating "is X
present / persisted?", runs a file-level git command to compare against HEAD:
git checkout analysis/SQL/foo.toml # or: git restore <path> / git stash
This reverts your uncommitted edit to that file back to HEAD. The agent then reads the reverted file, finds your change gone, and reports a false "missing / not persisted / blocker" finding. In a multi-agent review, other agents reading the same now-reverted file repeat the phantom — so a majority can "fail" on something that was actually correct and committed-in-spirit.
This is distinct from its siblings:
concurrent-session-checkout-clobbers-shared-worktree— an unknown session flips the branch (git checkout <branch>/git switch). Here the clobberer is an agent you dispatched running a file-level checkout/restore/stash.subagent-read-stale-worktree-needs-head-pin— read agents return stale data from the wrong worktree (no mutation). Here a single shared worktree is mutated underneath you.
Context / Trigger Conditions
- A review/verification agent's finding says a change you made is "missing", "not in the file", "not persisted", "appears in git diff but not the working tree".
- The agent's own evidence mentions running
git checkout/git restore/git stashor "git diff vs file content mismatch". - A
git statusyou run afterward shows a file you edited is no longer modified (it silently reverted), often with a harness "file was modified, either by the user or by a linter" reminder. - Decisive tell: your test suite that requires the supposedly-missing change passed before you dispatched the agents. Passing tests that need X prove X was present; the agent's git op reverted it after the fact.
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.
- 11d ago First seen · 164 lines · 409 tokens per session scan A 2f5dc8c7f31c
dispatched-bash-agent-git-checkout-clobbers-uncommitted-edit is a skill published in the GitHub repository wan-huiyan/agent-traffic-control (3 stars, last pushed 6d ago), licensed MIT. It adds 409 tokens to every session and 2,330 once invoked, about $0.0020 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 2026-08-31.
Other skills, from other repositories
cloudflare-workers-ci-cd
Complete CI/CD guide for Cloudflare Workers using GitHub Actions and GitLab CI. Use for automated testing, deployment pipelines, preview environments, secrets management, or encountering deployment failures, workflow errors, environment configuration issues.
claude-code-bash-patterns
Claude Code Bash tool patterns with hooks, automation, git workflows. Use for PreToolUse hooks, command chaining, CLI orchestration, custom commands, or encountering bash permissions, command failures, security guards, hook configurations.
bump-sdk-version
Bump the FutureSearch SDK version across all files. Use when releasing a new SDK version, updating version numbers, or the user says bump version, release, version bump.
pr-check
PR review compliance: fetch review comments from an open PR, categorize as resolved/unresolved/dismissed, critically evaluate fixable items, implement approved fixes, and reply inline. Pass --unattended to halt on human-judgment items (emitted as Pending-Human) instead of prompting via AskUserQuestion.
babysit
PR babysitter: monitors CI status, auto-rebases when behind, auto-fixes CI where possible, delegates review comment handling to dlc:pr-check, and re-requests review after fixes. Designed for /loop usage with Remote Control.
git-ops
Git hygiene: clean up merged branches, prune stale remotes, and verify repository state.