ghcli-mcp

ghcli-mcp is a skill for Claude Code from FerhatDundar/ghcli-mcp-connector. It costs 137 tokens per session (1,094 once invoked), scanned A, original, MIT.

A guide for performing GitHub tasks with the gh command-line tool, including checking pull requests, issues, releases, workflow runs, and account details.

In plain words
What is it for?
Running repository operations such as listing or inspecting pull requests and issues, checking workflow status, creating issues, merging changes, and identifying the logged-in account.
Why use it?
It standardizes GitHub commands, requires an explicit repository, and defines when actions need extra safety confirmation.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the ghcli-connector plugin — 1 skill, 1 MCP server shipped together

Good fit Running repository operations such as listing or inspecting pull requests and issues, checking workflow status, creating issues, merging changes, and identifying the logged-in account.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/ferhatdundar/ghcli-mcp-connector/ghcli-mcp
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 FerhatDundar/ghcli-mcp-connector --skill ghcli-mcp
Clone the repo
git clone --depth 1 https://github.com/FerhatDundar/ghcli-mcp-connector

Made for: Claude Code.

Or install ghcli-connector, the plugin that ships this one along with the rest of its 1 skill, 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 ghcli-mcp

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

Your own site · 80×15
<a href="https://agentmods.dev/skills/ferhatdundar/ghcli-mcp-connector/ghcli-mcp"><img src="https://agentmods.dev/badge/skills/ferhatdundar/ghcli-mcp-connector/ghcli-mcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 137 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,094 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 $0.00137 $0.01094
Opus 5 $0.00068 $0.00547
Sonnet 5 $0.00027 $0.00219
Haiku 4.5 $0.00014 $0.00109

Measured 8d ago against content hash 892a52aac173, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

ghcli-mcp 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 8d 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.

plugin/skills/ghcli-mcp/SKILL.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ghcli-mcp

Use the ghcli_exec, ghcli_help, and ghcli_whoami tools for any GitHub operation. They shell out to the real gh CLI on the host, so every gh command is available — not a hand-picked subset.

Tools

  • ghcli_whoami — show the authenticated GitHub identity. Call this first if it's unclear whether auth is even working, or the user asks "who am I logged in as".
  • ghcli_help — show gh <command> --help text. Call this before ghcli_exec whenever unsure of exact flag names or syntax — it's always safe, never gated.
  • ghcli_exec — run any gh <command> <subcommand> ... command. Pass the arguments as an array, without the leading "gh", e.g. args: ["pr", "list", "--repo", "owner/repo"].

Always pass -R/--repo explicitly

This connector's process has no working-directory git repo, so it can't infer which repo a command targets the way an interactive gh session would. Every repo-scoped command needs an explicit -R owner/repo or --repo owner/repo. If a user names a repo in conversation, use it; if they don't and the command needs one, ask which repo rather than guessing.

Safety model — read-only by default

Commands that change state on GitHub (create, delete, edit, close, merge, comment, reopen, archive, and gh api calls with --method other than GET) are blocked unless the connector's server has GHCLI_MCP_ALLOW_WRITE=true configured and the tool call passes confirm=true.

  • If a write is attempted while the server is in read-only mode, ghcli_exec returns a clear error saying so — relay that to the user rather than retrying with different arguments; the fix is a config change on their end (see the connector's SETUP.md), not a different command.
  • If the server does have writes enabled, still confirm with the user in plain language before calling ghcli_exec with confirm=true on anything destructive or hard to undo (closing/merging a PR, deleting something, editing a title/body) — the tool's confirm flag is a mechanical gate, not a substitute for actually checking with the person you're acting on behalf of.
  • Local-only side effects — repo clone, pr checkout, release download — are never gated, since they don't touch anything on GitHub. No need to ask before those.

Read the full file on GitHub · 108 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. 8d ago First seen · 108 lines · 137 tokens per session scan A 892a52aac173

Subscribe to this mod's changes

ghcli-mcp is a skill published in the GitHub repository FerhatDundar/ghcli-mcp-connector (0 stars, last pushed 2d ago), licensed MIT. It adds 137 tokens to every session and 1,094 once invoked, about $0.0007 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.