Use cases
Pyrite's typed entries, plugin system, and AI integration adapt to different knowledge domains. Here's how teams are using it.
Software teams
Architecture Decision Records, component documentation, prioritized backlogs, coding standards, and runbooks — all as structured, validated entries in git. Your AI pair programmer reads the project KB through MCP and has full context on every decision, every component, every backlog item.
# Your AI pair programmer knows your architecture
$ pyrite sw adrs
16 Architecture Decision Records
$ pyrite sw components
24 documented services and modules
$ pyrite sw backlog --status planned
Prioritized, effort-estimated, dependency-mapped
Pyrite itself uses this workflow — its own backlog, ADRs, and component docs live in a Pyrite KB. Dogfooding all the way down.
Research & investigations
Track people, organizations, events, and relationships across hundreds of sources. Timeline visualization shows when things happened. The knowledge graph reveals connections between entities that aren't obvious from any single document. Every claim links back to a source.
# Structured investigation
$ pyrite create --type person --title "Jane Doe" \
--body "CFO at Acme Corp since 2023. Previously at Initech."
$ pyrite search "financial connections" --mode hybrid
$ pyrite timeline --from 2023-01 --to 2024-06
Personal knowledge management
Zettelkasten-style note-taking with a maturity workflow: fleeting notes become literature notes become permanent notes. Wikilinks connect ideas. Semantic search finds conceptually related notes you forgot you wrote. AI helps surface connections and suggests links.
# Capture, connect, develop
$ pyrite create --type zettel --title "Thinking in Systems" \
--body "Meadows argues that system behavior..." --tags reading
$ pyrite search "feedback loops" --mode semantic
# Finds related notes across all your reading
Team wikis
Replace Notion or Confluence with something you own. Structured entries with wikilinks, collections with kanban/table/gallery views, a web UI your whole team can use, and AI features powered by bring-your-own-key. Deploy on a $6/month VPS with Docker — unlimited users, no per-seat pricing.