Skip to content

Repository files navigation

whodar - know who knows

whodar

Know who knows.

CI Latest release License

Someone at your company already knows the answer. whodar tells you who. Point it at the tools your org already uses, ask in plain words, and get the people to talk to and the channels to ask in, each with the reason and confidence behind it. Local by default, with or without an LLM.

See it

Ask from the terminal and get people, channels, reasons, and confidence:

whodar in the terminal

Or serve the local web UI, where every result carries a confidence badge and feedback buttons, a query lives in the URL so answers are shareable, clicking a person shows everything whodar knows about them, and a sidebar browses the whole graph: people, channels, teams, and topics:

whodar web UI

Install

brew install kordloom/tap/whodar

Or go install github.com/kordloom/whodar@latest, or grab a prebuilt binary from the releases page.

Verifying a download

Every release ships a checksums.txt signed with cosign. The signature is checksums.txt.bundle and the public key is cosign.pub in this repo. Download all three alongside your archive, then:

cosign verify-blob \
  --key cosign.pub \
  --bundle checksums.txt.bundle \
  --new-bundle-format \
  checksums.txt

shasum -a 256 -c checksums.txt --ignore-missing

The first command proves the checksum file is the one we signed; the second proves your archive matches it.

Quickstart

No data yet? Explore a simulated company across all eight sources, no credentials needed:

whodar demo

Then index something real:

whodar index --source org-csv --file examples/people.csv
whodar ask --pretty "who do I talk to about billing retries"

Then wire in the rest of your tools. The guided way is whodar connect, a wizard that explains each source, reads the token without echoing it, validates it, and runs the first index:

whodar connect

Prefer copy-paste? Every source has a recipe in docs/CONNECT.md, with the exact credential to create, the command to run, and how to verify it worked: slack, github, jira, confluence, pagerduty, git, and codeowners.

How it works

Piece What it does
Sources Eight pluggable connectors feed one graph of people, teams, topics, and channels. Adding a source is one small interface.
Identity One human stays one node: sources join by email, and an alias file joins handle-only identifiers like a GitHub login.
Ranking Owners beat chatterboxes: repetition saturates while explicit signals stay strong. Recency counts, every answer carries a confidence, and results explain which words hit where.
Feedback Confirm or correct a result and future rankings move, without burying the evidence.
Modes Keyword needs no model and always works; semantic and LLM answers run on local Ollama, or on Claude, Gemini, and OpenAI behind explicit opt-in.
Frontends The CLI, web UI, Slack bot, and an MCP server for agents like Claude Code all share one engine.

Data governance

Indexed work data is sensitive, so whodar controls what a model can see. The default policy is strict: answers are computed locally and nothing is sent to any model beyond this machine. The redacted policy admits only the known cloud providers (Claude, Gemini, OpenAI) and sends them your question plus numbered candidates, meaning title, team, and matched query terms, never names, emails, channel names, or message text. The open policy sends full candidate detail anywhere you point it. Indexing talks only to the sources you name, with your own tokens, and the index on disk is readable only by your user. Serving the web UI beyond localhost requires a bearer token on every request. An organization can pin the policy with a locked file that user flags and environment variables cannot override.

Docs

License

Business Source License 1.1 (see LICENSE). Source-available: run whodar in production inside your company on your own infrastructure, with no seat count and no license key. The one restriction is offering whodar to third parties as a hosted service that provides its primary functionality. It converts to Apache 2.0 on 2030-08-03.

Hosting it for others, embedding it in something you sell, or bound by a policy that rejects source-available licenses? See COMMERCIAL.md.

Copyright 2026 KordLoom LLC.

About

Expertise locator for teams: find who to talk to about X across Slack, GitHub, docs, and tickets. Self-hosted, local by default, works with or without an LLM.

Resources

Contributing

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages