Fetching from the wire…
01
02
03
04
05
06
07
08
OSS2026-07-25 · source-backed
OneCLI, Apache-2.0, attacks a problem every agent builder has: agents need API access, and embedding live secrets in each agent multiplies exfiltration surface. Agents get placeholder keys; a Rust gateway on port 10255 intercepts outbound HTTP, matches host and path patterns against an AES-256-GCM-encrypted store (PostgreSQL via Prisma), and injects the real credential. "Agents never see the secrets." Ships a Next.js dashboard on port 10254, scoped per-agent access tokens, and Bitwarden integration. 2.8k stars, 307 commits. This is the most concrete answer yet to a threat model where the failure was always a live secret sitting somewhere readable.
Each link below shares sources, entities, or timing with this story.
Encrypts your .env secrets so AI agents literally cannot read them. AES-256-GCM + Argon2id encryption with intentionally no get or export commands — secrets go in but can't come out to an agent. Brand new (Feb 22) and solving a universal pain point as agents gain more filesyst...
MIT-licensed alpha. Each "coworker" gets a real browser with its own logins, its own files and only granted tools, with every action decided before it happens and recorded after. Runs on Docker Compose against your own PostgreSQL, no model ships in the box, admin credentials e...
Block's buzz jumped to 10,934 stars (+2,506 in a day) as a self-hostable Apache-2.0 relay built on Nostr where every message from a human, agent, workflow, or git event is a signed event in one log. Agents get their own keypairs and audit trails and participate as first-class...
Triple-stream retrieval (BM25 keyword, vector embeddings, knowledge-graph traversal) fused via Reciprocal Rank Fusion on the iii engine, with SQLite for state and an in-memory vector index, no external database. The economic claim: ~170K tokens/year (~$10) versus ~650K tokens...
macro-inc/macro hit #4 on GitHub trending at 2,821 stars total, shipping email, chat, docs, tasks, calls and CRM that all @-link into a shared context graph agents read as team-level memory. Explicitly "fully open source, not open core," with commercial licensing sold separate...
MemPalace (~53.6K stars) claims best-benchmarked, against mem0 (~57.8K) and claude-mem (~80.8K) (GitHub). Agent memory went from experimental nicety to a competitive subcategory with published benchmarks. Pair this with the local-first angle: Mnemo offers a Rust + SQLite + pet...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.