shared-memorylisted

Search, save, and query a three-tier semantic memory shared across all AI agents on your workstation. Use before starting any task (search first for prior context) and after completing significant work (save findings with entities for consolidation). Supports save_decision for full PROV-O provenance and save_retrospective to record whether decisions held up — closing the Why-To loop.
KanenasInGreece/Shared_Memory · ★ 1 · AI & Automation · score 64

Install: claude install-skill KanenasInGreece/Shared_Memory

# Shared Memory (Hive-Mind) ## Overview This skill bridges the Shared Memory Framework — a three-tier semantic and relational memory layer shared across all AI agents on your workstation. Facts saved by one agent are retrievable by all others. Knowledge persists across sessions and tools. **Agents currently integrated:** | Agent | Skill invocation | Install path | |---|---|---| | Claude Code | `/shared-memory` | `~/.claude/skills/shared-memory/` | | Grok | `/shared-memory` | `~/.grok/skills/shared-memory/` | | Codex CLI | `$shared-memory` | `~/.codex/skills/shared-memory/` | | Antigravity CLI (`agy`) | `/activate shared-memory` | `~/.gemini/skills/shared-memory/` | | LM Studio | MCP `rag-orchestrator` | `vector-skill.py` via `mcp.json` | > **This skill is the usage surface — a thin client.** It runs one script, > `memory_bridge.py`, over HTTP to the gateway on `:8888`. It does not run or > manage the gateway, the daemons, or schema migrations — that **operations > surface** lives on the gateway host in the framework repo > ([Documentation/server-setup.md](Documentation/server-setup.md)). Installing > this skill is not installing the framework. --- > **AI instruction — use absolute paths for every CLI command.** Skill runners execute commands from the user's project directory, not the skill directory, so a bare `scripts/memory_bridge.py` fails with "No such file or directory." Commands below use the Antigravity CLI path (`~/.gemini` — the legacy directory name it inherit