Call-graph review context
Context Goblin builds a call graph of your codebase, detects the symbols changed in a pull request, and follows their callers in other files so reviewers see the affected blast radius.
Context Goblin
Context Goblin is an AI code reviewer for GitHub pull requests that adds call-graph and cross-repository context before writing a review. It helps teams catch issues that depend on how services, files, and external knowledge sources are connected.
Overview
Context Goblin is an AI code reviewer for GitHub pull requests. It focuses on the change in context, not just the diff, by combining the pull request with call-graph data, cross-repository relationships and other connected knowledge sources before it writes a review.
The product is aimed at teams that want automated review comments with deeper repository awareness. It can run from the dashboard, respond to pull request mentions, or be called through an MCP client, and it posts a GitHub review with a verdict, summary and inline comments.
Features
Cross-repository dependency awareness
It maps routes, queues, shared tables and package dependencies across repositories, then adds only the relations touched by the change to the review context.
Four review agents
The review is split across four isolated specialists for architecture, security, logic and tests, which keeps findings focused by concern.
Verified findings only
Every finding is checked against the codebase before being posted, and the system filters out style-only noise.
MCP context connections
Admins can connect Jira, Confluence, Notion, custom documentation or any MCP-compatible server, then choose which tools the agents may call.
Multiple review entry points
Reviews can be started from the dashboard, by commenting @contextgoblin on a pull request, or by calling the review_repository MCP tool.
Use Cases
Reviewing changes with wider system impact
Use Context Goblin when a pull request changes code with downstream callers, shared dependencies or cross-service effects, and you want the review to consider that blast radius before merge.
Checking cross-repository dependencies
Use it in a multi-repository setup where routes, queues, tables or package links span several services and reviewers need to see how the change propagates across repos.
Fitting into different review workflows
Use the dashboard, GitHub mention or MCP client to trigger a review in the workflow that fits the team, whether reviewers live in GitHub or work from another agent.
Adding non-code context to reviews
Use the MCP server connections when a code change depends on ticket context, design notes or internal docs stored in Jira, Confluence, Notion or another MCP-compatible source.
Publishing actionable review feedback
Use the structured verdict, severity labels and inline comments when you want a review output that can be acted on directly in GitHub instead of copied from a separate tool.
Pros and Cons
Pros
- Adds call-graph and cross-repository context to pull request reviews.
- Can incorporate external context through MCP servers such as Jira, Confluence and Notion.
- Separates review work into four focused agents for architecture, security, logic and tests.
- Provides a GitHub review state and inline comments rather than only a plain-text summary.
- Offers multiple ways to trigger a review, including dashboard, GitHub comments and MCP.
Cons
- The pricing page was not accessible from the collected sources, so plan details are unclear.
- Some setup steps are admin-only, including repository connections, MCP server configuration and certain settings.
FAQ
How do I start a review?
It reviews pull requests in connected GitHub repositories. You can start a review from the dashboard, mention @contextgoblin on a pull request, or use its MCP tool from your own client.
What context does Context Goblin use in a review?
The docs say it reads pull request diff, metadata, commits and discussion, then adds blast radius context, cross-repo routes and crawled relations before running four review agents.
What kinds of issues does it look for?
The built-in review agents focus on architecture, security, logic and tests. The coordinator combines their findings into a single review with a verdict and inline comments.
Can it use context beyond the codebase?
Yes. You can connect your own MCP servers, and the review agents can read context from tools such as Jira, Confluence and Notion, or any other MCP-compatible server.
Is there pricing information available?
The pricing page was not available, but the homepage shows a free start with one full crawl, first pull request reviews, and GitHub App and MCP server access. It also says reviews pause when free credits run out and nothing is charged automatically.
Quick Facts
- Category
- Developer Tool
- Platform
- GitHub pull requests
- Primary workflow
- AI review using codebase context and cross-repo relations
- Entry points
- Dashboard, @contextgoblin GitHub comment, MCP client
- Integrations
- GitHub App, MCP servers, Jira, Confluence, Notion
- Pricing
- Free start shown; paid plan details unavailable
Alternative a Context Goblin
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
Manta AI
Manta AI is an autonomous web app testing tool for teams that want to map application behavior, catch regressions, and generate tests without writing scripts or maintaining selectors. It works from a URL and supports plain-English test flows, run results with screenshots, and scheduled or deployment-triggered checks.
Ghost
Ghost è un assistente AI da terminale per chattare, generare codice ed eseguire task da riga di comando. Include modelli gratuiti, supporta Linux, macOS e Windows, ed è open source.
Vi3ecode
Vi3ecode is a maintained development environment for working with AI coding agents. It keeps project context, terminal, Git, memory, flows, review, and communication connected around the agent you already use.
Lucid Train
Lucid Train is a local-first AI coding harness that turns repository code into architecture diagrams and uses those diagrams as specifications for coding agents. It is available as a desktop app and a lightweight Rust CLI, and it can run with local models or your own API keys.
Hoplite
Hoplite is a cloud coding agent platform for teams that want to run software tasks in isolated sandboxes with repository context and imported local setup. It offers published Pro, Scale, and Enterprise plans.