openloomi-install
OpenLoomi install & first-use setup helper for Codex. Use when the user wants to install OpenLoomi, configure it, or troubleshoot `INSTALL_REQUIRED` / `SOURCE_FOUND_APP_NOT_BUILT` / `SESSION_INITIALIZATION_REQUIRED` after running setup-status. Triggers: install openloomi, configure openloomi, setup openloomi, openloomi not installed, openloomi not finalized, install_required, install missing, guest session.
git clone --depth 1 https://github.com/melandlabs/openloomi /tmp/openloomi-install && cp -r /tmp/openloomi-install/plugins/codex/skills/openloomi-install ~/.claude/skills/openloomi-installSKILL.md
# OpenLoomi Install Sub-skill This sub-skill is auto-loaded when the user wants to install or fix OpenLoomi on their machine. It composes `loomi-bridge` operations and never downloads or executes anything outside the plugin's own scripts. ## Quick workflow 1. **Always run `setup` (and the loopback checks below) outside the Codex sandbox** — the install + launch path needs writes to `/Applications`, desktop GUI launch, and `http://localhost:3414`. Default `read-only` / `workspace-write` modes will silently fail. Ask the user to drop `sandbox=workspace-write` (or move to `danger-full-access`) before invoking the bridge. 2. `node "$SKILL_DIR/../../scripts/loomi-bridge.mjs" setup-status` 3. Based on `nextAction` / `reason`: | Reason / nextAction | Action | |---|---| | `install_openloomi` / `INSTALL_REQUIRED` | OpenLoomi Desktop is not on this machine. **Call `setup --yes` to run the end-to-end wizard in one invocation**: resolve → download → install → set `OPENLOOMI_AGENT_PROVIDER=codex` → launch the desktop app → wait for the local API → mint a guest session token. Invoking this skill counts as explicit approval to pass `--yes`. For offline / corporate-proxy installs, point the user at the restricted-network guide (link below) and call `install-instructions` first. | | `SOURCE_FOUND_APP_NOT_BUILT` | A source checkout is present but the OpenLoomi Desktop GUI app has not been built yet. Recommend either building the source per the OpenLoomi repo's Tauri build docs (or run `loomi-bridge install-instructions`) or installing the packaged Desktop release. | | `open_openloomi` / `OPENLOOMI_API_UNREACHABLE` / `SESSION_INITIALIZATION_REQUIRED` | OpenLoomi is installed but the local API or guest/session token is not ready. Ask the user to open OpenLoomi Desktop once, or run `setup` so the bridge can launch/init through OpenLoomi-owned surfaces. | | `inspect_codex_runtime` | OpenLoomi is up but the native Codex provider is not the active default agent. Have the user confirm the Codex CLI is installed (`codex --version`) and rerun `setup` so the runtime env + LaunchAgent get re-applied. | Before applying the API-unreachable row, inspect `loopbackAccessAmbiguous`. When it is `true`, the result is inconclusive because the current Codex sandbox may block host loopback traffic. Request approval and run `loopbackAccess.verification.commands` outside the sandbox. Only tell the user to restart OpenLoomi if the outside-sandbox checks also fail. ## Sandbox and installation Downloading and installing OpenLoomi can require capabilities that are blocked inside the current Codex sandbox: - GitHub release lookup and artifact download require network access. - The default installer may write to a system application directory such as `/Applications`. - Interactive installers and launching OpenLoomi Desktop require GUI/process access outside the sandbox. When an install command fails with a likely network, permission, filesystem, or GUI sandbox error, request approval and retry the same `loomi-bridge` command outside the sandbox. Do not work around the bridge by downloading or installing the artifact with unrelated commands. Do not report that the release URL is invalid or the installer is broken until the approved retry also fails. ## Codex runtime setup If the user is setting up OpenLoomi from Codex, explicitly asks to switch the desktop runtime executor to Codex, or diagnostics show a native-agent provider mismatch, call: ```bash node "$SKILL_DIR/../../scripts/loomi-bridge.mjs" codex-runtime-info ``` ## Reminder: secrets contract - The bridge never reads AI provider env vars and never reports provider readiness from `/api/preferences/ai` — it only checks whether the native Codex runtime is active via `/api/native/providers`. Provider configuration is the OpenLoomi runtime's job, surfaced through the desktop app's own settings UI. - If the user pastes a key into chat, redact it: do NOT echo it back, and tell them to remove it from chat history. - Never pass `--api-key` as an argv flag. The bridge has no such flag, by design. - The bridge's end-to-end wizard is `setup --yes` (one-shot: install + set codex provider + launch desktop + mint guest session token). Invoke it from this skill when the user asks to install or finalize OpenLoomi. **Do not call `install-openloomi --confirm` by itself** — that only installs and leaves the desktop app unlaunched and the session token unminted. Use `install-openloomi --confirm` only when the user explicitly asks to install without launching (rare).
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.
Use 1000+ external apps via Composio - either directly through the CLI or by building AI agents and apps with the SDK
Walk the user through installing the optional composio skill and turning on screen memory
Persist learning across turns by actively maintaining MENTAL_MODEL.md in the workspace. Use for iterative coding, debugging, benchmarking, or any feedback-driven task; read and update the file with actual filesystem writes before every response, never just hidden/internal memory.
Create an end-to-end Continual Learning Bench task. Use when adding a new src/tasks/<name> benchmark task, including design interview, concrete Jinja prompt templates, JSON action contracts, observation templates, debug schedule, behavioral E2E tests, validation, smoke/debug runs with icl and icl_notepad, and README documentation.
Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. "open a file in TextEdit", "navigate to /Applications in Finder", "click the Save button in Numbers").
High-performance data analysis using Polars - load, transform, aggregate, visualize and export tabular data. Use for CSV/JSON/Parquet processing, statistical analysis, time series, and creating charts.
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of "Word doc", "word document", ".docx", or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a "report", "memo", "letter", "template", or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.