signetai
Persistent local memory, secrets, and knowledge graphs for coding agents
All Skills in This Repository (9)
Pure Emerald Level Indicatorssignet-design
Standardize Signet UI visuals with design tokens, typography, and icons.
signet
Provide persistent memory and identity for AI agents across sessions.
memory-debug
Diagnose and fix Signet memory issues across daemon health, embeddings, and data integrity.
recall
Query persistent memories using hybrid vector and keyword search.
skill-creator
Create new Claude Skill templates with SKILL.md structure and Python scripts.
web-search
Orchestrate SearXNG, Lightpanda, and Agent-Browser for web search and scraping.
agent-architect
Create SOUL.md, IDENTITY.md, USER.md, and AGENTS.md seed files defining agent voice and behavior.
remember
Persists selected content across Signet agent sessions with tagging and importance metadata.
onboarding
Guide users through an interactive onboarding interview to create local Signet identity and configuration files.
Frequently Asked Questions
FAQPage SchemaHow to install Signet?โผ
Run `npx skills add Signet-AI/signetai --all -g -y` in your terminal to install all skills in this suite globally.
What does Signet do for AI coding agents?โผ
Signet gives agents persistent memory across sessions by automatically extracting facts from transcripts and injecting relevant context into every prompt. It works across Claude Code, Codex, OpenClaw, Gemini CLI, and other harnesses through one shared knowledge base.
Does Signet store my data locally?โผ
Yes. Everything lives in a local SQLite database under ~/.agents/ on your machine, and nothing is sent to external servers. Secrets are stored in an encrypted vault and never appear in logs or transcripts.
Can I import my existing notes and documents?โผ
Yes. Signet imports Obsidian vaults, Discord, and GitHub in real time, plus one-time imports of Word, Excel, PowerPoint, PDF, EPUB, and CSV files into the memory graph.
How accurate is Signet's memory recall?โผ
Signet scores 97.6% average answer accuracy on the LongMemEval benchmark, and you can verify recall quality yourself using the built-in MemoryBench benchmarking skill.
Related Repositories in Software Engineering
View All in Software Engineeringโopenclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core