SignetAI | Self-Hosted AI Memory for Agents

Docs Blog Join Community<br>GitHub

PORTABLE AGENT STATE<br>███████╗██╗ ██████╗ ███╗ ██╗███████╗████████╗<br>██╔════╝██║██╔════╝ ████╗ ██║██╔════╝╚══██╔══╝<br>███████╗██║██║ ███╗██╔██╗ ██║█████╗ ██║<br>╚════██║██║██║ ██║██║╚██╗██║██╔══╝ ██║<br>███████║██║╚██████╔╝██║ ╚████║███████╗ ██║<br>╚══════╝╚═╝ ╚═════╝ ╚═╝ ╚═══╝╚══════╝ ╚═╝ ███████╗██╗ ██████╗ ███╗ ██╗███████╗████████╗<br>██╔════╝██║██╔════╝ ████╗ ██║██╔════╝╚══██╔══╝<br>███████╗██║██║ ███╗██╔██╗ ██║█████╗ ██║<br>╚════██║██║██║ ██║██║╚██╗██║██╔══╝ ██║<br>███████║██║╚██████╔╝██║ ╚████║███████╗ ██║<br>╚══════╝╚═╝ ╚═════╝ ╚═╝ ╚═══╝╚══════╝ ╚═╝

Identity, memory, and secrets that survive the model switch.

Signet keeps agent state outside any single model or harness. It preserves the exact record, builds useful context from it, and keeps credentials out of the model window.

$ curl -fsSL https://signetai.sh/install.sh | bash

Join Discord<br>GitHub<br>same binary via npm install -g signetai

PORTABLE CONTEXT ACROSS THE ECOSYSTEM Claude Code OpenCode OpenClaw Codex Oh My Pi pi

Agent Overview Skills Marketplace Knowledge Graph

CAPABILITIES 02 / 04<br>FIG_01 // AMBIENT_FLOW<br>ACTIVE_STREAM

Ambient Memory<br>Every session is automatically distilled into structured memory. No tool calls, no manual saves — your agent wakes up knowing what happened yesterday.

FIG_02 // KNOWLEDGE_GRAPH

Knowledge Graph<br>Entities and relationships extracted with high precision. Graph traversal allows for deep context retrieval.

FIG_03 // INSPECTION_UI

Inspect Everything<br>Memory is not a black box. Browse, search, and edit through a refined dashboard with full provenance.

FIG_04 // CROSS_HARNESS<br>Cross-Platform<br>A unified agent state layer that keeps identity and memory coherent across every major AI agent harness.<br>Claude Code OpenCode OpenClaw Codex

FIG_05 // CRYPTO_LAYER

Encrypted Secrets<br>Industry-standard encryption at rest. Agents utilize credentials without ever exposing plaintext.

FIG_06 // MULTI_TENANT

Multi-Agent<br>Granular visibility controls. Define isolated, shared, or group memory scopes for swarms.

FIG_07 // CONTEXT_RANKING

Context Ranking<br>Inspectable ranking, provenance, and dampening keep useful context visible without flooding the model window.

FIG_08 // EXTENSIONS Core Ecosystem

SDK & React Hooks CLI Interface Git Provider Sync Skills Marketplace Docker / Self-Host MCP Support

BENCHMARK 03 / 04

Early results on the LoCoMo long-conversation benchmark. Only local-first, open-source system in the top 10.

87.5% Answer Accuracy LoCoMo long-conversation benchmark, 8-question full-stack sample<br>100% Hit@10 Retrieval Every relevant memory found in the top 10 candidates<br>0.615 MRR Mean Reciprocal Rank — how high the right answer appears<br>LOCAL No LLM at Search Time Retrieval and ranking run entirely on-device

Larger evaluation runs in progress. These numbers will change.

Methodology and full results &rarr;

GET STARTED 04 / 04<br>Your agent has memory in five minutes.<br>01 Install<br>$ curl -fsSL https://signetai.sh/install.sh | bash<br>same binary via package managers: npm i -g signetai or bun add -g signetai

02 Setup<br>Run signet setup, pick your harness, and the daemon starts automatically. Hooks are installed. Identity files are created.

03 Work<br>Open your coding agent and start a session. Memory extraction happens automatically. Next session, context is already there.

Apache 2.0 &middot; Open Source &middot; Local-first<br>Join Community Quickstart docs GitHub