specstoryai/getspecstory

Install our local first extensions for your favorite AI IDE or Terminal Agent. Process your histories into reusable skills with Lore. Sync your conversations to the cloud. File issues and requests.

SpecStory – Capture & Search Your AI Coding Conversations

What it is

  • A local‑first CLI (and IDE extensions) that automatically records every interaction you have with AI coding assistants (e.g., Cursor, GitHub Copilot, Claude Code, Codex, etc.) and stores them in a hidden .specstory/history/ folder inside each project.
  • The recorded sessions can be synced (optionally) to the SpecStory Cloud, where they become searchable, shareable, and can be queried with a chat‑style interface.
  • A companion tool called Lore can mine those histories to generate reusable “agent skills” that encode the patterns you actually use.

Key features

  • Automatic capture – no manual copy‑paste; the CLI wraps supported agents and saves every prompt, response, and code snippet.
  • Local‑first privacy – data never leaves your machine unless you log in and explicitly sync.
  • Cross‑tool support – works with dozens of terminal agents (Claude Code, Cursor CLI, Codex, Droid, DeepSeek, etc.) and IDE extensions for Cursor and GitHub Copilot.
  • Search & chat – full‑text search across all saved sessions locally or in the cloud; ask natural‑language questions to retrieve past decisions.
  • Team sharing – optional cloud workspace lets teams share knowledge bases.
  • Lore skill generation – turn historic sessions into evidence‑backed agent skills that can be installed on any supported agent.

How to get started

  1. Install the CLI (Homebrew):
    brew tap specstoryai/tap
brew install specstory
  2. Run an agent through SpecStory – e.g., specstory run claude or specstory run cursor. The tool launches the underlying agent and auto‑saves the session.
  3. Sync to cloud (optional):
    specstory login   # create an account
specstory sync    # push local history
  4. Search or chat via specstory search <term> or by opening the web UI at cloud.specstory.com.
  5. Generate lore:
    npx skills add specstoryai/getspecstory --skill lore
# then invoke /lore inside a supported agent

Supported agents (partial list) – all work with the same CLI wrapper:

  • Cursor IDE & CLI
  • GitHub Copilot (VS Code)
  • Claude Code
  • OpenAI Codex CLI
  • Droid CLI
  • DeepSeek TUI
  • Antigravity CLI
  • Muse Code, Qwen Code, Pi, Grok Build, OpenCode, etc.

Typical use cases

  • Never lose a solution – revisit a bug‑fix or design decision months later.
  • On‑board new team members – share curated sessions that explain project conventions.
  • Build personal AI assistants – use Lore‑generated skills to give agents built‑in knowledge of your codebase and workflow.
  • Audit AI‑generated code – keep a searchable audit trail of what the model suggested and when.

Documentation & support

License

  • The core SpecStory CLI is open‑source (MIT‑style) and lives under specstory-cli/. The cloud service is a proprietary SaaS offering, but usage is optional.

Made with ❤️ by SpecStory

Related

  • Project
    dollspace-gay/chainlink

    A local-first issue tracker CLI that provides a persistent memory layer and behavioral guardrails for AI-assisted development to prevent context loss across sessions.

  • Project
    GeminiLight/MindOS

    MindOS is a local-first knowledge base that gives you and your AI agents a shared, persistent memory. It stores your knowledge as plain Markdown files, connects agents via MCP and Skills, and helps you turn reviewed conversations into reusable SOPs so agents start from a better baseline.

  • Project
    legeling/PromptHub

    A local-first workbench for managing and distributing AI prompts, skills, and programming rules across multiple AI coding tools like Cursor and Claude Code.

  • Project
    MemPalace/mempalace

    MemPalace is a local-first memory system for AI assistants (Claude Code, Codex, Cursor, any MCP client). It stores conversation histories verbatim — no summarization — and retrieves them via semantic search from a structured index organized into "wings" (people/projects), "rooms" (topics), and "drawers" (content). Pluggable vector backends (ChromaDB default; SQLite-exact, Rust-native, Milvus, Qdrant, pgvector optional) keep everything on-device by default, with no API calls needed. It ships a CLI, a 47-tool MCP server, a temporal knowledge graph, auto-save hooks, and Docker images. Claims 96.6% raw retrieval recall (R@5) on LongMemEval without any LLM, with reproducible benchmarks and unusually honest methodology notes.

  • Project
    FutureUniant/WorkShadow

    A local-first desktop work log application that uses AI for semantic search, automated report generation, and Q&A over personal work history while keeping data on the user's machine.