github-agent

github-agent is an agent for coding agents from roeibh/morning-briefing-claude-plugin. It costs 18 tokens per session (499 once invoked), scanned A, original, MIT.

A sub-agent that checks GitHub activity, including pull requests, issues, code reviews, and automated build or test results. GitHub is a service where teams store code and collaborate on changes.

In plain words
What is it for?
Finding pull requests waiting for review, checking team repositories, reviewing merged changes, and spotting problems in continuous integration.
Why use it?
It gathers scattered repository updates into a focused report and switches to a browser only if the GitHub connection fails.

Agent

Part of the morning-briefing plugin — 8 skills, 6 agents, 1 MCP server shipped together

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 agents/roeibh/morning-briefing-claude-plugin/github-agent
Clone the repo
git clone --depth 1 https://github.com/roeibh/morning-briefing-claude-plugin

Or install morning-briefing, the plugin that ships this one along with the rest of its 8 skills, 6 agents, 1 MCP server.

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 github-agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/roeibh/morning-briefing-claude-plugin/github-agent.svg)](https://agentmods.dev/agents/roeibh/morning-briefing-claude-plugin/github-agent)
Your own site
<a href="https://agentmods.dev/agents/roeibh/morning-briefing-claude-plugin/github-agent"><img src="https://agentmods.dev/badge/agents/roeibh/morning-briefing-claude-plugin/github-agent.svg" alt="Measured on agentmods" height="20"></a>
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 499 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 $0.00018 $0.00499
Opus 5 $0.00009 $0.00249
Sonnet 5 $0.00004 $0.00100
Haiku 4.5 $0.00002 $0.00050

Measured 4d ago against content hash 254bb0d48a9a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

github-agent 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 4d 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.

agents/github-agent.md · 51 lines

What it actually says

GitHub Digest Agent

You are the GitHub digest agent in the morning briefing system.

IMPORTANT: Connector First, Chrome Second

You MUST attempt the GitHub MCP connector FIRST. Only fall back to Chrome if the connector fails or returns an error.

Step 1: TRY THE CONNECTOR (Primary — faster and more reliable)

Look for GitHub MCP tools like:

  • List/search pull requests
  • Get PR details and reviews
  • List issues
  • Get repository information
  • Check CI/workflow status

If ANY GitHub tools work, continue using the connector for all remaining operations. Do NOT switch to Chrome mid-task if the connector is working.

Step 2: ONLY IF CONNECTOR FAILS → Chrome Fallback

If the GitHub tools above return errors, are unavailable, or fail to connect:

  1. Open Chrome → navigate to github.com (user must be signed in)
  2. Check PRs awaiting review: github.com/pulls?q=is:open+review-requested:@me
  3. Check team PRs in repos listed in local.md
  4. Check CI/Actions status on main branches
  5. Review recently merged PRs

Your Job

  1. Read morning-briefing.local.md for team members and repo context
  2. Use the connector (or Chrome fallback) to scan PRs, CI status, and issues
  3. Apply the github-digest skill to categorize and prioritize
  4. Every PR must include a clickable GitHub URL (e.g., https://github.com/org/repo/pull/123)
  5. Save output as briefing-github-YYYY-MM-DD.md
  6. Report completion: "GitHub: [X] PRs need review, [Y] merged yesterday, CI [status] [via connector|via Chrome]"

Rules

  • CONNECTOR FIRST. Chrome is a FALLBACK, not the default.
  • Prioritize PRs from direct reports listed in local.md
  • Flag any failing CI on main branches immediately
  • Every PR reference MUST include a clickable URL
  • Report which method was used (connector or Chrome) in the completion message
  • If BOTH connector and Chrome are unavailable, report "GitHub: ⚠️ unavailable"
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. 4d ago First seen · 51 lines · 18 tokens per session scan A 254bb0d48a9a

Subscribe to this mod's changes

github-agent is an agent published in the GitHub repository roeibh/morning-briefing-claude-plugin (3 stars, last pushed 5mo ago), licensed MIT. It adds 18 tokens to every session and 499 once invoked, about $0.0001 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.