Codna — repository intelligence runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 24 Sep 2026.
Map a repo before spending a token: triage, bug-fix PRs, SARIF reachability, on-device recall.
Today is the operative word: we check Codna — repository intelligence every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add codna -- uvx codna
{
"mcpServers": {
"codna": {
"args": [
"codna"
],
"command": "uvx"
}
}
}
[mcp_servers.codna]
command = "uvx"
args = ["codna"]
{
"mcpServers": {
"codna": {
"args": [
"codna"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"codna": {
"args": [
"codna"
],
"command": "uvx"
}
}
}
This one needs environment variables set before it will start:
ANTHROPIC_API_KEY (Provider key for codna_fix's planner (or store one with `codna key set`). Every tool also requires the one-time free `codna login` device authorization; codna_triage, codna_secure and codna_recall then need no key at all.), GITHUB_TOKEN (Write token, only for codna_fix with open_pr=true (opens a pull request) and codna_report_bug (files an issue).), CODNA_API_KEY (Codna key from the one-time free `codna login` (device authorization, free community license). Required to execute any of the five tools; introspection (initialize/tools/list) needs none.).
The author declared them in the registry entry; get the values from the project itself.
Repo intel for AI coding agents: overview, PRs, contributors, hot files, CI, deps. Remote MCP.
Session memory for coding agents: local-first recall, drift detection, on-device inference.
PageRank-curated code intelligence for coding agents over an indexed source repository.
Linux package, file, command, vulnerability, lifecycle, migration, and repository intelligence.
Analyze repos of any size - security scanning code analysis monorepo support
Cross-repository semantic code intelligence for AI coding agents, from a local code graph.
A local intelligence layer for AI coding agents. Maps your repos, matches the open-source frontier.
Repository evidence for agents before they adopt dependencies, enter codebases, compare, or merge.
Answers built from our own checks of this server.