Queries Sourcery security findings, updates issues, and builds remediation prompts through MCP or CLI.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent — or use 1-click editor setup below.
We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag — we're steadily working through the catalog.
💡 Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Sourcery.
sourcery_security_snapshotcounts by status/severity plus the first page of active issues — start here for a triage overview.
sourcery_list_findingslist issues with `repository_ids`, `issue_types`, `statuses`, `search`, `limit`, `cursor`.
sourcery_get_findingfetch one issue (full record incl. source snippet and dependency graph).
sourcery_get_security_countsaggregate counts by status and severity.
sourcery_bulk_update_findingsbulk status/severity changes (≤100 ids; `SOLVED` is scanner-owned and rejected).
sourcery_build_fix_promptbuilds the Sourcery-style minimal-change agent prompt from a finding; the DEPENDENCY path uses `fixed_versions` + `manifest_file_path` and renders the dependency chain.
The sourcery MCP server exposes Sourcery’s documented Team-plan REST API for security findings to MCP-compatible agents. Its scope is security issue and issue-group data rather than Sourcery’s pull-request review controls. The bundle also includes a command-line interface and skills for triage and remediation workflows.
The package is published as @abyssbugg/sourcery. It can be used directly as an MCP process or installed as a plugin bundle for supported agent hosts. A Python implementation is included as a reference, while the production server is built with the official MCP TypeScript SDK.
The server uses Sourcery’s API at https://api.sourcery.ai/api and authenticates with a bearer token supplied through SOURCERY_API_KEY. Its implementation is pinned to a committed OpenAPI snapshot. The snapshot defines eight verified API operations covering issue and issue-group listing, statistics, retrieval, and bulk updates.
The MCP layer presents those operations as focused tools. A snapshot tool combines status and severity totals with the first page of active issues, making it a starting point for triage. List operations accept filters such as repository IDs, issue types, statuses, search text, limits, and cursors. The fix-prompt tool creates a minimal-change prompt from a finding; dependency findings use fixed versions and the manifest path to render the dependency chain.
Node.js 20 or newer is required. For the default stdio mode, set the API key and start the package without arguments:
Streamable HTTP mode is available with http and listens on 127.0.0.1:8765/mcp:
The repository contains host-specific setup instructions for Claude Code, Cursor, Codex or ChatGPT desktop, VS Code, Devin, Grok, Amp, Hermes, OpenClaw, and Rovo Dev CLI. ChatGPT web requires the HTTP server to be exposed behind HTTPS and added through developer mode. The bundle can also be run from a local plugin directory.
The sourcery MCP server provides tools for both individual findings and issue groups:
The command-line interface offers equivalent workflows such as snapshot, list, get, and fix-prompt. Query commands accept --json for scripting. SOLVED is scanner-owned and rejected by bulk finding updates.
Sourcery’s public API is documented as a security-findings API. Review actions such as review, summary, guide, title, resolve, dismiss, and create issue remain GitHub or GitLab comment and label commands, so they are outside this server’s REST surface and should be handled through a Git provider connector.
The OpenAPI snapshot is pinned in the repository and includes a recorded fetch date and SHA-256 value. If Sourcery changes its specification, the snapshot and corresponding constants must be regenerated before relying on the new operations or enum values. The HTTP mode also requires suitable HTTPS exposure when used by ChatGPT web.
Factual signals from GitHub, npm, and our automated checks — not a rating.
No reviews yet — be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/sourcery-2)<a href="https://allmcps.com/mcp/sourcery-2"><img src="https://allmcps.com/api/badge/sourcery-2?style=directory" alt="Sourcery on AllMCPs" /></a>