pssah4

Deep ingest of a source (PDF/Markdown/URL/DOCX/PPTX/XLSX) in Karpathy multi-turn pattern. Forced Markdown conversion, block-refs to the source rendered as a discreet ↗ symbol. Mandatory step 1 is triage (cluster match, source diversity, tension hint).

Quick single-pass ingest of a source (PDF/Markdown/URL/DOCX/PPTX/XLSX) into one note (frontmatter + overview + key points + original text). PDFs stay page-refs (no mirror enforcement). Block-refs to the source rendered as a discreet ↗ symbol per key point. No multi-turn dialog.

Integrate new notes into the knowledge graph. Create PDFs as markdown source notes. Set properties, links, MOC entries. Prefer existing entities, propose new ones as stub notes.

Translate an Anthropic-style skill (with Python scripts) into a native Vault Operator skill (with JavaScript scripts that run in the sandbox). Use when the user asks to translate, convert, port, or import an Anthropic skill, e.g. "Hol mir den pdf-Skill von Anthropic" or "translate this python skill". A dry-run pass classifies every Python import against the mapping table; partial or unmappable cases trigger a user-confirmation modal before any file is written.

Guide for creating effective Vault Operator skills. Use when the user wants to build a new skill (or refine an existing one) that captures a repeatable workflow, a domain expertise, or a tool-integration recipe. Triggers include phrases like "build me a skill", "kannst du das automatisieren", "create a skill", "neuer skill", "make this repeatable".

Compact summary of a transcript note that stays digestible in under a minute. Strictly on the transcript, no interpretations. Sets block-IDs at key passages and links each statement in the summary discreetly via ↗ symbol to the source passage. Single-note layout (summary on top, transcript below).

Batch-rename attachments (PDFs, images) by convention. Author-Year_Title for sources, context-based for images.

Autonomous batch mode for Vault Health findings. Works through orphans, missing backlinks, and tags in batches without asking on every fix.

Convenience wrapper for existing DIA users upgrading between DIA versions (e.g. v1 -> v2 -> v3). Runs the migration scripts under `tools/migration/` in sequence: detect state, strip status duplicates, rename IDs (FEATURE-NNNN -> FEAT-EE-FF, EPIC-NNN -> EPIC-NN), flatten analysis/, regenerate BACKLOG.md, update skill-name references, run `/consistency-check` mode A. Idempotent. Use this skill when the user already has DIA artefacts and wants to upgrade to current conventions without the full reverse-engineering walk. Triggers: "migrate to DIA v2", "migrate v1", "upgrade DIA", "convert backlog", "restructure backlog", "FEATURE to FEAT". For BROWNFIELD projects (no `_devprocess/` yet, only code), use `/reverse-engineering` instead -- it absorbs the same migration mechanics for any pre-existing partial artefacts and additionally walks the code to fill the gaps.

Activates, reconfigures, or deactivates the Digital Innovation Agents workflow in a project. Writes the .dia/config.toml settings file and manages anchor blocks in agent-facing files (CLAUDE.md, AGENTS.md, GEMINI.md, .cursorrules, .github/copilot-instructions.md, .windsurfrules). Use this skill when the user mentions "DIA aktivieren", "DIA installieren", "DIA initialisieren", "DIA setup", "DIA settings", "Workflow-Modus", "Modus wechseln", "DIA off", "DIA deaktivieren", "Plugin aktivieren". Single skill that covers both first-time activation and later reconfiguration; the skill detects which path applies from the presence of .dia/config.toml.

Brownfield entry point for the V-Model workflow. Reverse-engineers an existing codebase into the standard V-Model artifacts: plan-context.md, ADRs, arc42 snapshot, FEATURE inventory, backlog seed, and an evidence-based BA draft. Walks the V backwards from Coding up to Business Analysis, filling every step only with what can be proven from the code or the existing documentation. Every claim is sourced (path:line or doc section); nothing is invented. Use this skill when the user mentions "existing project", "legacy codebase", "reverse engineer", "import existing code", "brownfield", "we already have code", "onboard existing project", or when the user wants to enter the V-Model workflow but artifacts do not exist yet.

Defines project structure, naming conventions, and way of working for all projects. Referenced by other skills to ensure consistent directory structures, file names, and documentation standards. Use this skill when the user mentions "project setup", "project structure", "conventions", "init", "initialize project", "directory structure", or similar. Also automatically relevant when starting a new project.

Guides users through the V-Model workflow: takes stock of the current project state, recommends the next phase skill, audits handoff entries for completeness, and runs the Closing Handoff after a successful security audit. The guide does not perform CRUD work on artifacts; every concrete change lives in the phase skill it dispatches to. TRIGGER when the user asks "where do I start", "what comes next", "I am lost in the workflow", "which phase fits", "wo bin ich gerade", "wo soll ich anfangen", "was kommt jetzt", or explicitly requests the V-Model overview or end-to-end orientation. DO NOT trigger for individual phase work (use the specific phase skill instead: business-analysis, requirements-engineering, architecture, coding, testing, security-audit), generic mentions of "workflow", "process", or "next step" inside an already-running phase skill, or when a phase skill is already active.

Verifies the V-Model artifact graph is intact: BA sections, Epics, Features, Success Criteria, ADRs, arc42 sections, PLANs, Backlog, Wayfinder rows, and code references. Two modes: syntactic (links, IDs, Refs) and semantic (content coherence via an agent). Default syntactic at the end of every skill phase. Semantic before release or on explicit request. Use this skill when the user mentions "consistency check", "graph check", "check references", "dead links", "orphan features", "graph-health", or when another V-Model skill completes a phase and wants to verify the artifact graph before handoff.

Handoff skill and bug-capture entry point: loads plan-context.md and all design artifacts, performs a critical review against the real codebase, and ensures continuous writeback to the artifacts during and after implementation. Also the entry point when a bug is reported outside of an active implementation run -- captures the FIX artefact (BACKLOG row + FIX detail file + branch) without forcing an immediate fix. Use this skill when the user mentions "implement", "code", "realize plan-context", "build feature", "Bug gefunden", "es gibt einen Fehler", "Fix erfassen", or similar, and either a plan-context.md / FEATURE spec exists OR a bug surfaced. This skill does NOT take over the coding workflow -- the Default Claude Code agent remains responsible for the actual implementation. This skill ensures the context is critically reviewed, cleanly handed off, and artifacts stay up to date.

Transforms business analyses into epics, features, and tech-agnostic success criteria. Creates handoff documents for the architect. Use this skill when the user mentions "Requirements", "RE", "Define Features", "Create Epics", "User Stories", "Requirements", "Success Criteria", "NFRs", "ASRs", "Acceptance Criteria", or similar. Also when a BA document exists and the next step is the formalization of requirements.