Semantic Code Intelligence
Vectorized codebase indexing and symbol search understand your architecture — not just keyword matches.
VS Code AI Extension
Codor brings multi-step task execution into VS Code. It reads files, runs terminal tests, and applies precise code edits with interactive diff reviews before saving.
From high-level request to verified code in three structured phases.
Codor semantically maps your files, AST symbols, and dependencies. It checks existing conventions before making changes.
Complex goals are broken down into real-time task checklists. The agent edits files, runs terminal tests, and self-corrects.
Interactive visual diffs let you inspect every modified line with instant revert and selective approvals before committing.
Vectorized codebase indexing and symbol search understand your architecture — not just keyword matches.
Interrupted turns, model switches, and IDE reloads resume seamlessly with full in-flight tool history.
Route tasks to leading frontier models with automatic fallback if upstream providers experience rate limits.
Review edits side-by-side with full syntax highlighting before changes touch your local filesystem.
Get started with autonomous coding for free.
For engineers shipping code daily.
For power users and complex codebases.
Need a custom enterprise contract? Contact sales
Traditional autocomplete suggests snippets as you type. Codor is a full autonomous engineering agent that breaks down multi-file features, searches symbols across your project, modifies files, runs terminal tests, and validates its work before you merge.
No. Your code is never used for training. Code indexing and tool execution run locally in your VS Code environment. Only the minimal context needed for the active prompt is sent securely over TLS.
Codor connects to top-tier frontier coding and reasoning models with automatic failover so your generation never gets blocked if a provider experiences rate limits.
Yes. You can run in interactive mode with manual approvals for file edits and shell commands, or enable autopilot for trusted tasks with real-time visual diff reviews.
Credits are metered per 1k input and output tokens at exact provider rates without opaque markups. Uninterrupted tasks and efficient history compression ensure zero token waste.