ThreadWeave captures the decisions, answers, and lessons scattered across your email, chat, and documents, and files them into a searchable palace. Self-hosted. PII-aware.
Most of what your firm knows is never written down where it can be found. It lives in email threads, chat messages, and people's heads. ThreadWeave captures it and files it into a palace where everything has its place.
One searchable home for everything your firm knows. Captured verbatim, never summarized away.
Engineering, sales, support. Every department gets its own wing in the palace.
Database, pricing, onboarding. A room per subject, so knowledge gathers where it belongs.
Each captured answer or decision is filed with its source, date, and author.
Related knowledge links within a team. Cross-team links become tunnels, so one team's solution surfaces in another team's problem.
Tacit knowledge flows in through the entrance. ThreadWeave files each capture into a wing and a room, and connects related knowledge with hallways and tunnels.
git clone https://github.com/PowerLooming/ThreadWeave
then
cd ThreadWeave && bash setup.sh
No new tools. No workflow change. ThreadWeave reads from the platforms you already use, and the knowledge inside them becomes part of your palace.
Google Workspace service account — polls inbox, threads replies, extracts decisions
Listens across spaces, skips bot messages, captures human answers and decisions
Google Docs, Sheets, Slides, PDFs, Markdown — extracts text, maps folders to teams
Graph connector push-syncs ThreadWeave entries into Copilot and Microsoft Search
# One-shot sync from Gmail (last 7 days) threadweave gws sync --source gmail --query "newer_than:7d" # Continuous watch mode threadweave gws watch --interval 300 # Harvest a departing employee's knowledge threadweave gws harvest --email lars@firm.com
Knowledge goes into the palace. Answers come out. Here's what that looks like in practice.
Runs on your hardware. No cloud dependency. No external API calls required. Built for firms that can't send data off-prem.
Public, internal, confidential, restricted, HR-privileged, client-confidential, legal-privileged. Auto-detected at ingest. Enforced at read.
Emails, phone numbers, personal identifiers are detected and filtered before storage. Your knowledge base stays compliant.
When someone leaves, drain their institutional knowledge — every decision, process, and client context from their email, chat, and documents.
Uses any OpenAI-compatible endpoint to classify content with nuance. Gracefully falls back to regex heuristics when no API key is configured.
Every capture is filed into a wing (team) and a room (topic), connected by hallways and tunnels to related knowledge. Verbatim, never summarized away.
Semantic similarity + org proximity + freshness + author authority. The right answer finds the right person.
Team membership with validity windows. Query what team someone was on in Q2 2023. Knowledge graph built on MemPalace's temporal triples.
Full OpenAPI docs at /docs. Every pipeline stage is a documented endpoint. Build your own integrations on top.
Every piece of content passes through four stages before it reaches the knowledge store.
# Classify text — is it worth saving? threadweave detect "We decided to use Postgres for JSONB support" # Output: { "should_save": true, "content_type": "decision", "confidence": 0.92, "has_pii": false, "suggested_title": "Decision: Use Postgres for JSONB support" }
Ingest anything. Search everything. The pipeline does the rest.
# Ingest content — pipeline handles dedup, detection, PII, storage POST /api/v1/ingest { "content": "Always run integration tests before deploying to prod.", "source": "manual", "metadata": { "wing": "engineering", "room": "deployment", "author_id": "harald" } } # Search across everything POST /api/v1/search { "query": "deployment pipeline", "wing": "engineering" } # Metrics in Prometheus format GET /api/v1/metrics/prometheus
Star the repo, spin it up, or get in touch if you want to run it at your firm. No license fees. No cloud dependency. MIT, forever.