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 justincampbell/gh-watch --skill watch-branchgit clone --depth 1 https://github.com/justincampbell/gh-watchWrote 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/justincampbell/gh-watch/watch-branch)<a href="https://agentmods.dev/skills/justincampbell/gh-watch/watch-branch"><img src="https://agentmods.dev/badge/skills/justincampbell/gh-watch/watch-branch/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/justincampbell/gh-watch/watch-branch"><img src="https://agentmods.dev/badge/skills/justincampbell/gh-watch/watch-branch.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 | $0.00045 | $0.00719 |
| Opus 5 | $0.00023 | $0.00360 |
| Sonnet 5 | $0.00009 | $0.00144 |
| Haiku 4.5 | $0.00005 | $0.00072 |
Grade A, and why
watch-branch 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 12d 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Watch a Branch
Monitor a branch for new commits using the gh watch CLI extension. It polls GitHub and emits one JSON event per line to stdout when the branch tip changes, which makes it a natural fit for the Monitor tool — each event arrives in the conversation as a notification.
Prerequisites
Check if the extension is installed:
gh watch --help
If not installed:
gh extension install justincampbell/gh-watch
Usage
Run with the Monitor tool (not Bash with run_in_background). The first notification is the initial-state snapshot; subsequent notifications are new-commit events as they land.
Wait for the next commit on main, then exit:
command:gh watch branch $ARGUMENTS --exit-on new-commitdescription:New commits on $ARGUMENTSpersistent:falsetimeout_ms:3600000(1 hour — raise/lower based on how long you're willing to wait)
Watch the branch for the rest of the session (every new commit reported as it lands):
command:gh watch branch $ARGUMENTSdescription:Commits on $ARGUMENTSpersistent:true(no timeout — stop withTaskStopor end of session)
Flags
| Flag | Description | Default |
|---|---|---|
--interval <duration> |
Polling interval | 60s |
--exit |
Exit after any state change | false |
--exit-on <events> |
Exit after specific event types (comma-separated) |
Event types
| Event | Description |
|---|---|
initial-state |
Snapshot of the branch tip at the moment watching started (always emitted first) |
new-commit |
A new commit was pushed to the branch |
Output format
One JSON object per line on stdout. The first line is always initial-state — a snapshot of the branch at the moment watching began:
{"timestamp":"...","event":"initial-state","summary":"Branch main at abc1234: Latest commit message","details":{"branch":"main","sha":"abc1234...","message_headline":"Latest commit message","author":"octocat"}}
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.
- 12d ago First seen · 76 lines · 45 tokens per session scan A 40ffa549219b
watch-branch is a skill published in the GitHub repository justincampbell/gh-watch (4 stars, last pushed 24d ago), licensed MIT. It adds 45 tokens to every session and 719 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 2026-08-31.
Other skills, from other repositories
bellwether
Bring the current PR to a mergeable state: CI green, all review comments resolved, no merge conflicts. Self-contained — watches CI, fixes issues, watches again until merge-ready. Use when the user wants to keep a PR green, auto-fix CI, resolve review comments, or says "get this merged".
cross-squad-communication
Protocol for sending queries, delegating tasks, and sharing context between independent Squad instances across different repositories.
developer-release
Release management, changeset workflow, firewall log parsing, breaking CLI change rules, and Go module summaries for gh-aw.
github-mcp-server
Reference for GitHub MCP server tools, methods, and usage patterns.
pr-finisher
Prepare an open pull request for merge from a GitHub Copilot cloud agent. Drives Reviews, local validation, and Mergeable to a ready state. Does not merge, and cannot trigger CI.
distributed-mesh
How to coordinate with squads on different machines using git as transport.