Manages GitHub Pull Requests for authoring, reviewing, and iterating on code changes. Supports creating PRs, listing open PRs, reading PR details, reading and posting comments (general and inline), requesting reviewers, merging, and closing PRs.
AI agents invoke github_pull_requests to trigger actions in SushiMCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool spans multiple categories. Merging PRs triggers external operations with significant side effects (code lands in main branch), closing PRs is a semi-destructive action, and posting comments/requesting reviewers are write actions. Merging is the most impactful — it executes a workflow that integrates code changes into a repository, which can have broad downstream consequences.
From the tool's definition Supports creating PRs, listing open PRs, reading PR details, reading and posting comments (general and inline), requesting reviewers, merging, and closing PRs
Documented attack patterns abuse exactly the kind of access github_pull_requests gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and SushiMCP, and nothing reaches the server without passing your rules. This is the rule we recommend for github_pull_requests:
{
"version": "1",
"default": "deny",
"tools": {
"github_pull_requests": {
"limits": [
{
"counter": "github_pull_requests_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}github_pull_requests stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Manages GitHub Pull Requests for authoring, reviewing, and iterating on code changes. Supports creating PRs, listing open PRs, reading PR details, reading and posting comments (general and inline), requesting reviewers, merging, and closing PRs. It is categorised as a Execute tool in the SushiMCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Sushi MCP server in PolicyLayer and add a rule for github_pull_requests: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches SushiMCP. Nothing to install.
github_pull_requests is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the github_pull_requests rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for github_pull_requests. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
github_pull_requests is provided by the Sushi MCP server (maverickg59/sushimcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from SushiMCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
7 SushiMCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.