github-platform

github-platform is a skill for Claude Code, Codex from monkilabs/opencastle. It costs 54 tokens per session (1,694 once invoked), scanned A, original, MIT.

A coding-agent guide for working with GitHub, a service for hosting code and coordinating changes. It covers pull requests, review discussions, failed Actions checks, and stacked pull requests, which are dependent changes arranged in a sequence.

In plain words
What is it for?
Use it to open or review pull requests, respond to review threads, inspect failed GitHub Actions jobs, and split related changes into a stack.
Why use it?
It gives the agent a defined way to read and update GitHub work while handling authentication, permissions, pending reviews, and CI failures.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to open or review pull requests, respond to review threads, inspect failed GitHub Actions jobs, and split related changes into a stack.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/monkilabs/opencastle/github-platform
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.

Any agent
npx skills add monkilabs/opencastle --skill github-platform
Clone the repo
git clone --depth 1 https://github.com/monkilabs/opencastle

Made for: Claude Code, Codex.

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-platform

README.md
[![agentmods](https://agentmods.dev/badge/skills/monkilabs/opencastle/github-platform/github.svg)](https://agentmods.dev/skills/monkilabs/opencastle/github-platform)
Your own site
<a href="https://agentmods.dev/skills/monkilabs/opencastle/github-platform"><img src="https://agentmods.dev/badge/skills/monkilabs/opencastle/github-platform/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.

agentmods 80×15 button for github-platform

Your own site · 80×15
<a href="https://agentmods.dev/skills/monkilabs/opencastle/github-platform"><img src="https://agentmods.dev/badge/skills/monkilabs/opencastle/github-platform.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 54 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,694 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00054 $0.01694
Opus 5.5 $0.00022 $0.00678
Sonnet 5.5 $0.00011 $0.00339
Haiku 4.5 $0.00005 $0.00169

Measured 7d ago against content hash d2f2f6c95d56, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

github-platform 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 7d 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.

src/orchestrator/plugins/github/skills/github-platform/SKILL.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GitHub Platform

Branch, commit and pull request conventions: the git-workflow skill. Docs: https://docs.github.com/en/pull-requests

The MCP server

GitHub's own remote server, https://api.githubcopilot.com/mcp/. VS Code signs in with OAuth. Every other assistant sends GITHUB_PERSONAL_ACCESS_TOKEN, so the server acts as that token's user — call get_me first to know whose name every comment will carry. Tool names change between releases: read the tool list the server offers instead of assuming a name.

  • Toolsets. On by default: context, repos, issues, pull_requests, users. An X-MCP-Toolsets header on the server entry turns on more (actions for CI logs), and X-MCP-Readonly: true drops every write tool.
  • Token. Fine-grained and limited to the repositories the agents work in: Contents, Pull requests and Issues read and write, Metadata and Actions read. A 404 on a repository you know exists is the token: its repository list, or an organization that has not approved it.
  • gh covers the rest: review threads, job logs and stacks, below. It signs in on its own (gh auth status).

Gotchas

  • A review is pending until submitted. Comments added to a pending review are invisible to everyone else, and a session that ends before the submit leaves them unsent.
  • Why a PR cannot merge: gh pr view 42 --json mergeStateStatus. BEHIND needs a rebase onto its base, DIRTY has conflicts, BLOCKED lacks a required review or check, UNSTABLE has a non-required check failing.
  • A person merges, as git-workflow says. Stop at a reviewed PR with green checks, though both the server and gh offer a merge.
  • Open a draft while work remains (gh pr create --draft) and mark it ready when it is done (gh pr ready 42). Closes #123 in the body closes the issue only when the PR merges into the default branch.
  • In a review comment, a ```suggestion block replaces the commented lines when the author applies it. Use it for any fix that fits those lines.
  • Search allows 30 requests a minute. Narrow with repo: and is:open rather than paging.

Read the full file on GitHub · 78 lines

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. 7d ago First seen · 78 lines · 54 tokens per session scan A d2f2f6c95d56

Subscribe to this mod's changes

github-platform is a skill published in the GitHub repository monkilabs/opencastle (80 stars, last pushed yesterday), licensed MIT. It adds 54 tokens to every session and 1,694 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-09.