maludb

Full-length timed mock exam for the Claude Certified Architect (CCA-F) certification. Use when the user says "mock exam", "practice exam", "exam simulation", "test me like the real exam", "full practice test", or wants exam-condition practice for the Claude certification.

Progress report for Claude Certified Architect (CCA-F) exam preparation. Use when the user asks "how am I doing", "am I ready for the exam", "show my certification progress", "study report", or wants to know what to study next for the Claude certification.

Interactive study session for the Claude Certified Architect (CCA-F) certification exam. Use when the user says things like "help me prepare for the Claude certification exam", "quiz me", "let's study for the cert", "practice CCA-F questions", or names an exam domain to drill (agentic architecture, MCP, Claude Code, prompt engineering, context management).

Integrate the MaluDB REST/JSON API into a desktop (Electron) or web application. Use this whenever the user is calling MaluDB over HTTP — building a client for subjects, verbs, episodes, statements, documents, notes, projects, memory pools, skills, typed attributes, the knowledge graph, or the vector-memory search pipeline; authenticating with a `malu_` Bearer token against `api.maludb.com` / `/v1/...`; handling MaluDB error codes; or writing fetch/axios/requests/cURL code that talks to any `/v1` endpoint. Trigger even when the user only says "the MaluDB API", "call the memory API", "upload a document to MaluDB", or names a `/v1` path, without saying "client". For writing SQL straight against the database use maludb-sql-client instead; for building the PHP API server itself use maludb-api-server.

Build or extend the MaluDB PHP API server — the JSON+HTTP layer that fronts the `maludb_core` PostgreSQL extension. Use this whenever the user is writing or modifying the server side of MaluDB: adding a new `/v1/...` endpoint, creating the one-file-per-endpoint PHP files under `v1/`, wiring `.htaccess` URL→file rewrites, using the shared `config/response.php` helpers (`require_auth`, `body_json`, `json_response`, `json_error`, `db_query`/`db_exec`/`db_one`, `db_tx_core`), implementing Bearer-token auth, SQL tracing, the error→HTTP-status mapping, or the multipart document upload. Trigger even when the user just says "add an endpoint to the MaluDB API", "the v1 PHP files", "the MaluDB server", or "edit response.php", without saying "API server". For calling the API from a client use maludb-api-client; for raw SQL against the DB use maludb-sql-client.

Query and write the MaluDB PostgreSQL extension (`maludb_core`) directly with SQL from middleware — no REST API in between. Use this whenever the user is writing SQL against MaluDB: SELECT/INSERT/UPDATE/DELETE through the `maludb_*` updatable views; calling facade functions like `register_svpor_subject`, `maludb_register_episode`, `maludb_svpor_statement_create`, `maludb_svpor_attribute_create`, `maludb_memory_search`, `maludb_graph_walk`/`_neighbors`, `maludb_upload_document`; using the `malu_vector` type; setting the `search_path` for facade access; dealing with RLS/tenant ownership, the provenance lifecycle, or PostgreSQL constraint/trigger errors from MaluDB. Trigger even when the user just says "the MaluDB database", "query maludb_core", "the maludb_subject view", or names any `maludb_*` object, without saying "SQL client". For calling the HTTP API instead use maludb-api-client; for building the PHP API server use maludb-api-server.

Use when a project has gone down a bad path (wrong stack, over-engineered, incomplete requirements) and the developer wants to start over simpler without losing what they learned. Recovers the real requirements and emits a portable spec bundle to rebuild. Also runs on session start to detect and accept an existing salvage/ bundle. Triggers on "start over", "salvage this project", "rebuild simpler".

Reference for choosing the simplest stack that meets a stated requirement. Use during the salvage architecture interview to frame each technical fork and record a rationale. Covers data store, frontend, realtime, auth, jobs, and deploy.

Frontend development patterns combining HTMX, Bootstrap 5.3, and Alpine.js for interactive web applications. Use when building user interfaces, creating interactive components, styling with Bootstrap, implementing HTMX interactions, adding Alpine.js reactivity, handling forms, managing UI state, or any frontend development with HTMX and Bootstrap.

PHP backend architecture and development patterns for HTMX applications using LAMP stack (PHP, MariaDB/MySQL, Apache). Use when working on an HTMX-based PHP project for creating new features, structuring endpoints and partials, implementing CRUD operations, adding authentication or security, organizing code, debugging server-side issues, or any development task involving the PHP backend with HTMX responses.

Load, find, and query agent skills stored in a MaluDB (maludb_core) database. Use when you need to register a SKILL.md bundle into the knowledge graph, search for an existing skill by subject/verb/keyword or semantic match, or fetch a skill's markdown and files so it can be used. Covers the maludb_skill_register / maludb_skill_search / maludb_skill_get facades, the SVPOR discovery-tag scoring model, visibility and search_path rules, and the background reindex that keeps tags fresh.