diskd-ai

Summarize lessons learned from ccbox session logs (projects/sessions/history/skills) so the agent can do better next time. Produce copy-ready instruction updates (project + global) backed by evidence, with optional skill-span context to attribute failures to specific skills. Use when asked to run /ccbox:insights, generate a "lessons learned" memo, or propose standing instructions from session history.

Inspect and summarize local Codex/Claude session logs using the ccbox CLI (projects, sessions, history). Use when asked to list discovered projects, find sessions for the current repo/folder, print a session timeline, or produce an evidence-based “what happened” report from a session .jsonl log under CODEX_SESSIONS_DIR.

Query `.belief_map.sexp` before non-trivial code changes to find module boundaries, dependencies, blast radius, and the minimal files to read. Use for architecture discovery, impact analysis, and boundary checks.

Create technical design documents for new features, systems, refactoring plans, API designs, and database schemas. Use when asked to write a design doc, TRD (technical requirements document), RFC, or system design document. Triggers on requests like "write a design doc for X", "design the Y feature", "create a technical spec", or "plan the architecture for Z".

Convert AutoCAD DWG drawings into multipage PNG previews, readable tiles, a lossless position-referenced text layer, and a searchable metadata index — for RAG/VLM ingestion and CAD content extraction. Use when working with .dwg files, rendering DWG to images, extracting text/labels/title-block attributes from CAD drawings or blueprints, building RAG or multimodal pipelines over engineering/construction/survey drawings, or running the dwg2png CLI/Docker tool. Triggers on DWG, AutoCAD, CAD drawing, blueprint, dwg2png, "DWG to PNG", CAD text extraction, or indexing engineering drawings.

Google Gemini API integration for building AI-powered applications. Use when working with Google's Gemini API, Python SDK (google-genai), TypeScript SDK (@google/genai), multimodal inputs (image, video, audio, PDF), thinking/reasoning features, streaming responses, structured outputs with JSON schemas, multi-turn chat, system instructions, image generation (Nano Banana), video generation (Veo), music generation (Lyria), embeddings, document/PDF processing, or any Gemini API integration task. Triggers on mentions of Gemini, Gemini 3, Gemini 2.5, Google AI, Nano Banana, Veo, Lyria, google-genai, or @google/genai SDK usage.

Install, configure, and troubleshoot `assemblyai-cli` (Rust) to transcribe audio/video into text/srt/vtt using AssemblyAI, including speaker diarization. Use in any agent/tool that supports Agent Skills (Codex, Claude Code, etc.), especially for turning local media files or YouTube links (via `yt-dlp`) into transcripts.

Implement and maintain integrations with Nebius Token Factory (tokenfactory.nebius.com) OpenAI-compatible API, including auth + base URL config, chat/completions/embeddings/images, model discovery, batch, files, fine-tuning, custom models, and datasets/operations. Use when adding a Nebius provider adapter/client in TypeScript/JavaScript/Python or configuring OpenAI-compatible tooling (OpenAI SDK, LangChain, LiteLLM, etc.) to target Nebius.

Manage Redmine issues/tickets from the command line via the Redmine REST API using a `red-cli`-style interface. Uses `~/.red/config.json` for auth (server + api-key) with optional local overrides in `./.red/config.json`. Use when you need to create issues (defaults to Task tracker), list issues/projects, view issues (with journals), add journal notes/comments (public or private), edit issue fields (subject, description, status, priority, tracker, parent, assigned-to), manage issue relations (relates, blocks, duplicates, precedes, follows), update/remove existing comments (journals) by journal id, filter/sort/paginate lists, or troubleshoot Redmine CLI configuration (project-id/user-id, multi-instance `--rid`). Triggers on Redmine, red-cli, issue create/list/view/note/edit/relate/comment, subtask, parent, relation, `--rid`, `.red/config.json`, `api-key`, `project-id`, `user-id`, or requests to automate Redmine from a CLI.

Anthropic Claude API integration for building AI-powered applications. Use when working with Anthropic's Messages API, Claude SDKs (Python or TypeScript), tool use/function calling, vision/image inputs, streaming responses, prompt caching, message batches, token counting, extended thinking, PDF processing, or any Claude API integration task.

Create, update, and troubleshoot Tilt configuration (Tiltfile, Tiltfile extensions, and common Tiltfile API usage like docker_build/custom_build/live_update, k8s_yaml/k8s_resource, local_resource, and Docker Compose/Helm workflows). Use when asked to add Tilt to a repository, adjust dev-environment orchestration, or debug Tiltfile evaluation/resource update issues. Prefer consulting the local Tilt documentation source at /Users/alexeus/src/tilt.build/docs (and the generated API includes under /Users/alexeus/src/tilt.build/src/_includes/api) when available.

Clarify redmine task skill

Apply and review classic desktop UI designs against the 1992 Human Interface Guidelines. Use when designing or auditing menus, windows, dialog/alert boxes, controls, icons, color, mouse/keyboard behaviors, language/messages/help (including Balloon Help), localization/worldwide compatibility, accessibility (universal access), and collaborative-computing UX for desktop-style interfaces.