Superpositional

What is Superpositional?

A code intelligence platform that indexes your repositories and gives you answers, reviews, and drift detection across the full engineering lifecycle.

Superpositional is a code intelligence platform. It indexes your repositories into an internal agentic datastore, then uses that understanding to answer questions, review pull requests, generate documentation, detect when docs go stale, and investigate issues — automatically.

What you can do with it

Ask questions about your codebase

Open the Chat page and ask questions in plain English. Architecture, dependencies, security, change history — Superpositional searches your indexed repositories and gives you answers grounded in the actual code, with references to the specific files and functions that informed the response.

You can also connect AI coding tools like Cursor, Windsurf, VS Code, and Kiro via MCP. Your agent gets structured context from your indexed codebase, not just the files you have open.

Get automated PR reviews

When a pull request is opened or updated, Superpositional reviews it automatically. It checks the changes against configurable guards — coding standards, architectural boundaries, failure handling, security patterns — and posts findings directly to the PR as a GitHub check run.

Detect stale documentation

When Superpositional generates a document, it extracts structured claims from it — specific, verifiable statements about the code. When code changes later, those claims are checked against the updated codebase. If a claim no longer holds, the document is flagged as stale. No LLM call required for the check — it's a structural comparison.

Investigate issues and failures

When a GitHub issue is opened or a CI workflow fails, Superpositional can investigate automatically. It traces through the indexed codebase to identify potential root causes and suggests fixes.

Generate documentation

Superpositional can generate repository summaries, changelogs, and architecture documentation from your indexed code. Generated docs are kept fresh by the staleness detection system described above.

How it works at a high level

  1. You connect your GitHub repos. Superpositional installs as a GitHub App and watches for pushes, PRs, issues, and other events.
  2. Your code is indexed. Repositories are cloned, analysed, and indexed into Superpositional's internal datastore. This happens automatically on first connection and re-runs when code changes.
  3. Events trigger actions. A push re-indexes the affected code. A PR triggers a review. An issue triggers an investigation. These pipeline runs are called flows and you can track their progress in real time.
  4. You query the results. Via the Chat page, via MCP in your editor, or via the generated documents and reviews that Superpositional produces automatically.

Main features

FeatureWhat it does
ChatNatural-language questions about your codebase
MCPStructured codebase context for AI coding tools
PR reviewsAutomated code review with configurable guards
Document generationRepository summaries, changelogs, architecture docs
Freshness detectionFlags stale docs when code changes invalidate them
Issue investigationRoot cause analysis triggered by GitHub issues
CI failure diagnosticsAutomatic investigation when workflows fail
GuardsConfigurable coding standards enforced on every PR
FlowsReal-time tracking of all pipeline runs
Slack integrationNotifications and queries from Slack

On this page