Concepts

FAQ

Why Roka instead of uploading to Claude?

File size: Claude limits uploads; Roka can take a local file of any size the OS can read. Cost: a huge log billed as LLM tokens is expensive; prune locally and send only the slice. Speed: local prune is typically hundreds of milliseconds, not a full upload round-trip. Local CLI keeps logs on the machine.

Why not only semantic search / RAG?

Search finds chunks. Roka also force-keeps crash lines and collapses identical repeats so 1,000 identical warnings become one counted line. Default MCP tools use regex + dedup + recency, not embeddings. Semantic re-rank is a Pro/optional path on the hosted product.

Does it work offline?

Yes for the Free CLI prune path. Pro API and MCP verification need the network.

Do you store my logs?

Local CLI: logs stay on your machine. MCP tools prune in-process. Privacy policy: we do not store, read, or transmit log file contents as a product feature; pruning is local or transient. Account/usage metadata is separate — see Privacy.

Which agents?

CLI stdout works with any model. MCP connect writes config for Claude Code, Cursor, Codex, Copilot, VS Code, Windsurf, and Cline.

Quotas

Free: 15K tokens lifetime per device. Pro: 500K / month. Enterprise: unlimited. Processing time on the marketing site is cited around 312ms for a ~50K-token dump; large files scale roughly linearly.

Production workflows

Yes — pipe CLI output in CI or have the agent call prune_file on incident dumps. Treat this as early-access software (see Terms).

Support

mukhamedjankydyrli@gmail.com · CLI issues · source for MCP: GitHub org Roka-Dev-Labs (package still published on npm as roka-mcp).