Sourcegraph Cody?

NOT REALLYA prototype of the interface is possible, but the durable product depends on difficult systems beyond the UI. Build a safer adjacent workflow instead.

BUILD FIRST

A non-sensitive or read-only version of the core experience
A manual review step wherever correctness or trust matters
A narrow companion tool that removes a real piece of work

LEAVE FOR LATER

Regulated decisions, payments, or irreversible actions
Claims of correctness, compliance, or uptime that the MVP cannot support
Sensitive integrations until security, support, and auditability are ready

RESEARCH NOTES

What this tool actually does.

Checked 2026-09-06

Sourcegraph Cody is an AI coding assistant that combines LLMs with deep codebase context. It uses Sourcegraph's Search API to pull context from local and remote repositories and integrates through IDE extensions for VS Code, JetBrains, and Visual Studio, plus code hosts like GitHub and GitLab. Core features include: chat with @-mention context chips for files, symbols, repos, and line ranges; single/multi-line autocomplete; auto-edit that suggests changes based on cursor position and recent typing; Smart Apply to insert chat code suggestions as diffs; terminal command execution; debugging help; image upload for vision-capable models; customizable, shareable prompts; and context filters to ignore selected repos. Users can switch between LLMs optimized for speed or accuracy. Cody reports which files it used to answer and handles multi-file questions like locating component definitions and

SOURCES

RESEARCHED BUILD PROMPT

Give this to your coding agent.

Build a replacement for Sourcegraph Cody: a local-first AI coding assistant that answers codebase questions and writes code using project context, not a chat-with-LLM toy. Workflow: user opens a workspace and connects a local Git repo; a backend indexes the repo (chunking by symbols and file structure); a chat panel accepts questions and @-mentions for files/repos/symbols, streams answers, and shows which files were used; the editor provides inline autocomplete and diff-style edits that the user can apply. Technical requirements: VS Code extension in TypeScript, a local Node/Python server for indexing and retrieval, embeddings plus keyword search, tree-sitter for chunking, Git integration for branch awareness, streaming responses via an OpenAI-compatible API, and a model switcher. Implement a context filter to exclude specified repos. Use the extension's diff/apply APIs for code edits. Do not copy Sourcegraph branding, UI, or proprietary search; design your own UI. Out of scope: multi-user enterprise server, remote code host indexing, training custom models. Include a README covering setup, architecture, and known limitations.

Visit the official product page