> **Warning:** This Epic's description is a mirror of the [Selora Homes Roadmap](https://selorahomes.com/docs/roadmap/selora-ai-skills-library/). Do not edit this description directly in GitLab. # Selora AI: Library of Skills ## Overview Home Assistant blueprints let users share and reuse automations. [OpenClaw skills](https://docs.openclaw.ai/tools/skills) let developers extend AI assistants with reusable capabilities. Selora AI Skills combine both ideas into a single concept: **a curated, community-driven library of AI behaviors purpose-built for Home Assistant**. Each skill is a self-contained package that teaches Selora AI how to handle a specific domain — from energy optimization to security routines to comfort scenes. Skills are digested and configured by the [embedded Claw](/docs/roadmap/selora-ai-embedded-chat/) running on each Selora instance, so they work locally, securely, and without cloud dependency. Anyone can contribute a skill. The community curates, reviews, and improves the library over time — just like Home Assistant blueprints or OpenClaw's skill ecosystem, but focused entirely on the smart home. ## Example > **User**: "I want my home to save energy when nobody's around." > > **Selora AI**: "There's a community skill called *Away Mode Energy Saver* that uses your presence sensors > and smart plugs to cut standby power when the house is empty. Want me to install it?" > > **User**: "Yes, set it up." > > **Selora AI**: "Done — I've configured it with your motion sensors and smart plugs. It'll activate after > 30 minutes of no presence. You can adjust the delay anytime." ## Customer value - **Blueprints for AI**: Like Home Assistant blueprints, but instead of static YAML automations, skills give Selora AI contextual intelligence — the AI understands *intent*, adapts to device availability, and handles edge cases conversationally. - **Community-driven**: Anyone can author and share skills. The library grows organically with real-world use cases contributed by homeowners, installers, and developers. - **No code required**: Users discover, install, and configure skills entirely through natural language in the Selora AI chat — no YAML, no templates, no dashboards. - **Local-first execution**: Skills are digested by the embedded Claw on each Selora instance, ensuring they run locally and respect the user's privacy. - **Hardware-aware**: Skills declare their device requirements, so Selora AI only suggests skills compatible with the user's actual setup. ## Scope (first iterations) - **Skill specification**: Define the format, metadata, and requirements for a Selora AI skill (device dependencies, configuration parameters, required integrations). - **Community repository**: A public, Git-based repository where contributors submit, review, and version skills — inspired by the Home Assistant blueprints exchange and OpenClaw's skill registry. - **Embedded Claw integration**: The [embedded Claw](/docs/roadmap/selora-ai-embedded-chat/) ingests skills from the library and makes them available to the local Selora AI instance. - **Conversational discovery & deployment**: Users browse, install, and configure skills directly from the Selora AI chat interface — one conversation, zero YAML. - **Versioning & updates**: Skills support semantic versioning; installed skills can be updated in-place when new versions are published. ## Architecture (directional) - Each skill is a structured package containing: metadata, device/integration requirements, prompt instructions, and optional configuration parameters. - The community repository is a public Git repository with CI-based validation (linting, schema checks, security review). - The embedded Claw periodically syncs the skill catalog and indexes available skills against the local device registry. - When a user requests a skill, the Claw resolves device compatibility, walks the user through configuration, and activates the skill locally. ## Target customers - **Homeowners**: Access advanced AI behaviors with a single conversation — no technical knowledge needed. - **Installers**: Deploy proven, standardized skill sets across client installations for consistent results. - **Developers & contributors**: Share expertise with the community and build reputation through a public contribution model. ## Open questions and risks - **Security & validation**: How to sandbox and review community-contributed skills before they reach user instances — especially skills that interact with security-sensitive devices (locks, alarms, cameras). - **Skill quality**: Establishing review criteria and a curation process to maintain library quality as contributions scale. - **Proactive recommendations**: Integration with the [Suggest Integrations & Devices](https://gitlab.com/groups/selorahomes/products/-/epics/51) feature to proactively recommend skills based on the user's hardware. - **Attribution & incentives**: Recognition or incentive models for active contributors. - **Offline catalog**: How to handle skill discovery and updates when the instance has limited or no internet connectivity. ## Related - Selora AI: Embedded Claw (Epic #45): [Roadmap](/docs/roadmap/selora-ai-embedded-chat/) — the runtime that digests and executes skills locally. - Selora AI: Suggest Integrations & Devices (Epic #51): [GitLab epic #51](https://gitlab.com/groups/selorahomes/products/-/epics/51) — proactive hardware recommendations that can trigger skill suggestions. - Selora AI: Proactive Automation Suggestions (Epic #47): [Roadmap](/docs/roadmap/proactive-automation-suggestions/) — complementary feature that suggests automations; skills provide the reusable building blocks. - Selora AI: Create Scenes (Epic #46): [Roadmap](/docs/roadmap/selora-ai-create-scenes/) — skills can bundle scene-creation capabilities for specific use cases.