# vestige

> Cognitive memory for AI agents — FSRS-6 spaced repetition, 29 brain modules, 3D dashboard, single 22MB Rust binary. MCP server for Claude, Cursor, VS Code, Xcode, JetBrains.

- **URL**: https://www.freshcrate.ai/projects/vestige
- **Author**: samvallad33
- **Category**: MCP Servers
- **Latest version**: `v2.1.23` (2026-05-28)
- **License**: AGPL-3.0
- **Source**: https://github.com/samvallad33/vestige
- **Language**: Rust
- **GitHub**: 480 stars, 45 forks
- **Registry**: github
- **Tags**: `ai-memory`, `claude`, `cognitive-science`, `cursor`, `embeddings`, `fsrs`, `long-term-memory`, `mcp`, `rust`

## Description

Cognitive memory for AI agents — FSRS-6 spaced repetition, 29 brain modules, 3D dashboard, single 22MB Rust binary. MCP server for Claude, Cursor, VS Code, Xcode, JetBrains.

## Recent releases

| Version | Date | Urgency | Changes |
| --- | --- | --- | --- |
| `v2.1.23` | 2026-05-28 | High | v2.1.23 hardens the Sanhedrin launch path so Receipt Lock is portable, observable, and precise enough for broader use.  ### Added - Model-agnostic Sanhedrin backend presets for OpenAI-compatible servers, laptops, Ollama, MLX, vLLM, llama.cpp, hosted APIs, and Anthropic via LiteLLM. - Fail-open telemetry in `fail-open.jsonl`, plus a dashboard telemetry API and 7-day ambient dashboard counters. - Receipt schema documentation covering receipt artifacts, appeals, command ledgers, fail-open logs, com |
| `v2.1.22` | 2026-05-25 | High | v2.1.22 makes the optional Sanhedrin hook quieter and more accountable by turning draft judgment into local, appealable receipts instead of opaque vetoes.  ### Added  - **Receipt Lock** blocks unsupported verification claims such as "tests passed"   unless the current transcript contains a matching successful test, build,   lint, or typecheck command receipt. - **Veto receipts** are written to `~/.vestige/sanhedrin/latest.json` and   `latest.html` with Claim -> Verdict -> Precedent -> Fix -> Ap |
| `v2.1.2` | 2026-05-07 | High | ## Honest Memory  v2.1.2 focuses on operational trust: exact search stays exact, purge really removes content, contradictions are directly inspectable, and the update flow no longer depends on copied curl commands.  ### Added  - **Concrete search mode** — `search` now auto-detects literal queries such as quoted strings, env vars, UUIDs, paths, and code identifiers. Those queries take a keyword/literal path that skips HyDE, semantic fusion, FSRS reweighting, retrieval competition, and spreading a |
| `v2.1.1` | 2026-05-01 | High | v2.1.1 focuses on user-controlled portability: exact storage archives, merge-safe file sync, pluggable sync backends, and explicit hook opt-ins.  ### Added  - **Exact portable archives** — `vestige portable-export` / `vestige portable-import` preserve raw Vestige storage rows: memory IDs, FSRS state, graph edges, suppression state, audit rows, sessions, intentions, and embedding blobs. - **Merge-safe imports** — `vestige portable-import --merge` can merge into non-empty databases. It applies `s |
| `v2.1.0` | 2026-04-27 | High | ## [2.1.0] - 2026-04-27 — "Cognitive Sandwich Goes Local"  The Sanhedrin Executioner — Vestige's veto layer for Claude Code responses — now runs entirely on a local MLX model (`mlx-community/Qwen3.6-35B-A3B-4bit`). Zero API cost per Claude turn, fully offline, no Anthropic round-trip on the critical path. Combined with four pre-cognitive UserPromptSubmit hooks (synthesis-preflight, cwd-state-injector, vestige-pulse-daemon, preflight-swarm), Vestige now ships a complete "Cognitive Sandwich" — Ves |
| `v2.0.8` | 2026-04-23 | High | # Vestige v2.0.8 "Pulse"  The Pulse release wires the dashboard through to the cognitive engine. Eight new dashboard surfaces expose `deep_reference`, `find_duplicates`, `dream`, FSRS scheduling, 4-channel importance, spreading activation, contradiction arcs, and cross-project pattern transfer — every one of them was MCP-only before. **Intel Mac is back** on the supported list (Microsoft deprecated x86_64 macOS ONNX Runtime prebuilts; we link dynamically against Homebrew `onnxruntime` instead). |
| `v2.0.7` | 2026-04-20 | High | Hygiene release plus two UI gap closures. No breaking changes, no schema migration affecting user data beyond V11 dropping two verified-unused tables.  ## ✨ Added  - **`POST /api/memories/{id}/suppress`** — Dashboard users can now trigger top-down inhibitory control (Anderson 2025 SIF + Davis Rac1 cascade) without dropping to raw MCP. Optional `{"reason": "..."}`. Each call compounds. Emits `MemorySuppressed` so the 3D graph plays the violet implosion. - **`POST /api/memories/{id}/unsuppress`** |
| `v2.0.6` | 2026-04-18 | High | Polish release aimed at new-user happiness. v2.0.5's cognitive stack was already shipping; v2.0.6 makes it *feel* alive in the dashboard and stays out of your way on the prompt side.  ## ✨ Added  ### Dashboard reacts to six live events (was: one)  Before v2.0.6, the 3D graph was silent on six real cognitive events — the live feed showed them firing but the graph stayed motionless, which made the dashboard feel broken during real work. Now every event has a visual reaction driven by the existing |
| `v2.0.5` | 2026-04-14 | High | **The first AI memory system that can actively forget.**  Vestige now treats forgetting as a first-class, neuroscientifically-grounded primitive. New `suppress` MCP tool applies top-down inhibitory control over retrieval — each call compounds a penalty (up to 80%), a background Rac1 worker fades co-activated neighbors over 72h, and it's reversible within a 24h labile window. **Never deletes** — the memory is inhibited, not erased.  Based on [Anderson et al. 2025](https://www.nature.com/articles/ |
| `v2.0.4` | 2026-04-09 | High | ## v2.0.4 — "Deep Reference"  The biggest feature release since v2.0.0. Vestige now **reasons** across memories, not just retrieves them.  ### New: `deep_reference` — Cognitive Reasoning Engine  An 8-stage pipeline that builds a **pre-built reasoning chain** the AI validates — no LLM call needed:  1. Broad retrieval + cross-encoder reranking 2. Spreading activation expansion (finds connected memories search misses) 3. FSRS-6 trust scoring (retention × stability × reps ÷ lapses) 4. Intent classif |

## Citation

- HTML: https://www.freshcrate.ai/projects/vestige
- Markdown: https://www.freshcrate.ai/projects/vestige.md
- Dependencies JSON: https://www.freshcrate.ai/api/projects/vestige/deps

_Generated by freshcrate.ai. Indexes github releases for AI-agent ecosystem packages._