This release adds 2 notable features for engineering teams evaluating rollout.
Published 12d
MCP Developer Tools
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
ai-agents
ai-memory
bm25
claude-code
codex
cursor
+13 more
knowledge-graph
lancedb
llm
local-first
mcp
memory
model-context-protocol
privacy
python
semantic-search
sentence-transformers
sqlite
vector-db
Summary
AI summaryUpdates Eval harness, reversible, and J-score across a mixed release.
Full changelog
Own-your-data CLI commands
Use PMB as a personal, offline memory you fully control. All of these are
display/write-layer only - retrieval recall@10 and ~70ms p50 are unchanged.
pmb timeline- chronological, day-grouped view of your memorypmb insights- analytics: totals, weekly growth, top topics, lessons/goalspmb digest [today|week|month]- recap of recent memoriespmb export [--format markdown|json]- dump all memory to readable textpmb forget-topic <topic>- archive everything about a topic in one command (reversible)pmb ttl <ulid> 30d+--ttlon note/learn/fact +pmb prune-expired- memory expirypmb tag/untag/tags/tagged- local tags & collectionspmb reminders- overdue / due-soon goalspmb snapshot create|list|restore- local, offline workspace snapshots
Eval harness
- LLM-as-judge (J-score) harness improvements: full reader context,
per-question failure logging, sharper reader prompt. Eval-only - no
recall/product code touched.
Install
pip install -U pmb-ai
Full changelog: CHANGELOG.md
What's Changed
- feat: local-use commands + J-score eval harness by @oleksiijko in https://github.com/oleksiijko/pmb/pull/7
Full Changelog: https://github.com/oleksiijko/pmb/compare/v0.2.3...v0.2.4
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About PMB
All releases →Related context
Related tools
Earlier breaking changes
- v0.9.0 Deletes `packs/ru.yaml` and `packs/uk.yaml`; default active packs set to empty tuple.
Beta — feedback welcome: [email protected]