AI Agent Workflow Collection

Use this skill EVERY TIME an aphorism is added, validated, corrected, or imported into the Bazodiac collection under knowledge/bazodiaac-brain/aphorisms/. Verifies source attribution against trusted sources, refines DE and EN translations against the semantic core of the original, enforces the bilingual aphorism schema exactly (YAML frontmatter, body sections, blockquote, controlled vocabulary, attribution status), recomputes word counts, and either delivers a clean .md file or sorts the entry out with a reasoned verdict when attribution is unclear or rights are fraught. Triggers on phrases like "neuer Aphorismus", "Aphorismus hinzufügen", "kurate diesen Spruch", "validiere dieses Zitat", "Block 2 / Block 3", "Aphorismus-Eintrag", "ist dieser Spruch von ...", "passt die Zuschreibung". Manual schema enforcement is error-prone — past four files contained eight independent schema violations.

Pre-deployment verification checklist.

Skill for implementing the 3-step Memory-First Protocol (Research, Response, Learning). Use when starting ANY task to understand past context, and MUST use when completing a task to save new conventions, architectural decisions, or learned user preferences to the memory/ directory.

Identify distinct software components from design artifacts and create per-component directories. Use when transitioning from Design to Code phase.

Draft or update design documents based on approved specification. Use when translating requirements into technical specifications. Covers architecture, data model, API design, and design completeness assessment.

Interactive requirements elicitation for the Specification phase. Use when defining, reviewing, or modifying stakeholders, goals, user stories, requirements, assumptions, or constraints. Also performs gap analysis on existing artifacts.

Apply a user-reported fix, bug correction, or change to one or more components. Gathers context interactively, identifies affected components, builds informative context from design artifacts and decisions, implements the fix with tests, and handles design gaps. Use during the Code phase or later.

Create a phased implementation plan from design artifacts. Populates tasks.md with short tasks grouped into incremental development phases, each ending with a deployable/testable system. Use after SDLC-decompose, before coding begins.

Guided project initialization for the SDLC scaffold. Use when setting up a new project or re-configuring an existing one. Sets the project description in CLAUDE.md so AI agents have the right context from the start.

Project-wide status dashboard across all SDLC phases. Use when the user wants an overview of the entire project state. Aggregates artifact counts, statuses, phase gate readiness, traceability health, and assessment freshness into a single report.