Menu
SkillsMP has collected 19 skills from Eckii24/dotfiles. Open a skill to review its source and details.
Showing 19 of 19 collected skills.
Fetch public URLs/articles/docs/blogs and save clean Markdown with source URL, headings, lists, links, and code blocks. Use when the source is a URL. Not for converting local files (use markitdown) or condensing/summarizing content (use the /summarize-article…
Initialize state-of-the-art greenfield Python projects with current stable Python, uv, uv_build, Ruff, ty, pytest, a src layout, locked dependencies, and CI. Use when creating a new Python library, packaged application, CLI, service, worker, or standalone…
Sharp professional roast-style critique for code, architecture, docs, UI, specs, plans, prompts, or ideas when user asks for brutal feedback. Not for actionable line-level code review (use code-review-excellence) or interactive plan interrogation (use…
Use when converting local PDFs, Office files, images, or documents to Markdown or structured text. Defaults to Docling; uses MarkItDown only for its distinct fallback formats.
Implement a bounded code change from a request, spec, or implementation plan. Use when executing implementation work and choosing between quick, bounded, and tracked modes. Not for producing a spec (use to-spec) or an implementation plan (use to-plan) before…
Collapses deterministic multi-step repository research, aggregation, indexing, validation, and transformation into one small purpose-built local script and one Bash run. Use when a task would otherwise require 6+ related read/search/tool calls, iterates over…
Process open .ai-review.json code-review comments in the current Git repository when the user explicitly asks to work through, resolve, or inspect those comments.
Review a concrete change against an explicit focus: plan/spec fidelity, correctness, security, performance, tests, maintainability, architecture, or full review. Use for PRs, branches, implementation output, and targeted risk reviews. Not for grilling a…
Diagnose a bug, failing test, regression, or performance problem before fixing it. Use when something is broken, failing, wrong, slow, or intermittently unreliable. Not for implementing a known bounded change.
Relentlessly interview the user to sharpen a plan, decision, or design. Use only when explicitly asked to grill, stress-test, or deeply clarify thinking. Not for code review, direct implementation, or a bounded question.
Create/review/update Presenterm or Markdown slide decks, talk outlines, speaker notes, themes, exports, and presentation flow/readability. Use when authoring or reviewing a slide deck or talk; not for prose documents or non-slide markdown.
Build throwaway prototypes. Terminal apps for logic/state questions, or toggleable UI variations for design exploration.
Turn a bounded functional spec into an implementation plan. Use when architecture, code structure, sequencing, migration, and verification must be decided before implementation. Requires an existing bounded spec; not for defining what to build (use to-spec)…
Turn current context into a concise implementation spec. Use when a user wants to define what to build, capture requirements, or create a bounded technical/product spec before implementation. Not for mapping an unclear or large initiative (use wayfinder),…
Map a large or unclear initiative into facts, decisions, open frontier, and next work without writing a spec or implementation plan. Use when scope, architecture, dependencies, or sequencing are still genuinely uncertain before creating a spec. Not for an…
Create, revise, or retire a reusable agent skill with accurate routing, progressive disclosure, and verified resources. Use when authoring or materially changing a SKILL.md, its description, references, assets, scripts, or reusable workflow guidance. Do not…
Locate and inspect other P0-owned repositories already cloned under the REPO_PATH directory when a dependency, package, service, or internal library name points outside the current repo. Use when the current repo references an internal P0 dependency and the…
Create a concise ADR for a durable, consequential architectural decision. Use when a significant, hard-to-reverse technical decision needs recording with context, alternatives, and consequences. Not for routine or easily reversible choices.
Manage tracked feature work with `.ai/current-work.md`, handoffs, and archive rules. Use when initializing, updating, handing off, or archiving multi-session tracked work. Not for one-off tasks needing no persistent state.