yazanabuashour

Use installed openclerk document and openclerk retrieval JSON. With explicit user content or runner-supported public-source context, let the agent/OpenClerk propose path, title, body preview, tags, fields, and next approved request; do not write before approval. Bootstrap no-tools rule - if required fields, content/body, or update target are missing and no faithful candidate, duplicate-risk, artifact, or public-link placement proposal can be formed; if numeric limit is negative; or if bypass is requested with SQLite, raw vault/file/repo inspection, HTTP/MCP, legacy/source-built paths, unsupported transports, backend variants, module-cache inspection, rg, find, ls, external OCR, browser automation, local file reads, or opaque artifact parsing, this description is complete. For those invalid cases only, do not open this skill file, run commands, use tools, or call the runner; respond with exactly one no-tools assistant answer naming the missing/invalid fields or unsupported workflow, and ask for missing fields.

Use the optional Tesseract OCR module only after it is installed and enabled, keeping OCR review local-first, read-only, provenance-bearing, and approval-before-write.

Use the optional Gemini embeddings module only after explicit opt-in, keeping remote semantic retrieval read-only, rate-limit-aware, redacted, and citation-bearing.

Use the optional Ollama embeddings module only when it is installed and enabled, keeping semantic retrieval explicit, local-first, read-only, and citation-bearing.

Use OpenBrief for local-first brief runs through the installed OpenBrief JSON runner. Reject direct SQLite reads or writes as runner JSON substitutes, HTTP/MCP/source-built runner bypasses, and unreviewed imports of delivery history, latest-seen state, run state, or inferred private configuration. User-pointed legacy automation/config migration may inspect only the named input, draft OpenBrief sources and outlet policies for review, and apply after approval through openbrief config.

Use this skill for local-first OpenHealth weight, body-composition, blood-pressure, sleep, medication, lab, or imaging data. Reject directly without tools or file reads for short dates without a year, year-first slash dates, invalid values, unsupported units/statuses, invalid lab slugs including slashes, empty optional text or notes, unsafe corrections/deletes, medication end before start, systolic not greater than diastolic, invalid sleep quality or wakeup count, or any mixed-domain request with an invalid write; do not run validate. For valid requests, use this skill's runner contract, assume openhealth is on PATH, and pipe JSON directly to the installed runner; never command -v, --help, search source/docs/module cache/SQLite, or inspect code. Batch same-domain rows in one runner call; after writes, answer from returned entries unless a different final filter is requested.