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 pr-followup-commit-stranded-after-squashgit 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/pr-followup-commit-stranded-after-squash)<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/pr-followup-commit-stranded-after-squash"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/pr-followup-commit-stranded-after-squash/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/pr-followup-commit-stranded-after-squash"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/pr-followup-commit-stranded-after-squash.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 | $0.00201 | $0.02221 |
| Opus 5 | $0.00101 | $0.01111 |
| Sonnet 5 | $0.00040 | $0.00444 |
| Haiku 4.5 | $0.00020 | $0.00222 |
Grade A, and why
pr-followup-commit-stranded-after-squash 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 9d 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 — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Follow-up commits stranded after PR squash-merge
Problem
You've been iterating on a PR: pushed commit 1, opened the PR, then pushed commits 2 / 3 / ... in response to user feedback or new information. At some point the user merges the PR — possibly during one of your iterations, possibly before you finished pushing. The squash on main only contains commit 1's content. Your follow-up commits sit on the closed branch, never reaching main. If you ask gh pr view N --json state you see MERGED, but the merged content is missing your iterations.
This is a silent failure. Nothing errors. The PR appears successfully shipped. But main is incomplete.
Context / Trigger Conditions
All of:
- You pushed multiple commits asynchronously to a single PR's branch (not all at once before opening the PR).
- You believed those commits would all land in the squash because they were on the branch at merge time.
gh pr view N --json statereturnsMERGED.git show <merge-commit> --statshows insertion counts matching only your first (or first few) commits — not all of them.git log --oneline <stranded-branch>..origin/mainreturns empty (the branch's later commits never reached main).- The closed feature branch still exists locally / on GitHub, holding the stranded commits.
Why squash captures less than you think
GitHub's squash-merge captures the state of the branch at the moment of merge. If the user clicks "Squash and merge" on the GitHub UI when only commit 1 is on the branch, that's what gets captured. Commits you push after that moment land on a closed branch — GitHub does NOT auto-rebase the closed branch's later commits onto main.
The trap is the mental model. "Squash collapses everything that ever lived on the branch" sounds right but it's wrong. The correct mental model: squash captures the commits that exist on the branch at merge time. Push timing matters.
Solution — recovery recipe
Once you've confirmed commits are stranded, recovery is cheap:
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.
- 9d ago First seen · 150 lines · 201 tokens per session scan A 64c528d8a7bf
pr-followup-commit-stranded-after-squash is a skill published in the GitHub repository wan-huiyan/agent-traffic-control (3 stars, last pushed 4d ago), licensed MIT. It adds 201 tokens to every session and 2,221 once invoked, about $0.0010 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.
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.
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.
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.