caneff
- 64 skills
- 0 followers
- 3 hours ago last updated
- ▌ Extract Coding Standards · caneffExtract the coding standards a repo actually enforces from recent PRs, reviews, commits, and current code, and write them into CODING_STANDARDS.md. Use when the user says "extract coding standards", "mine recent commits or PRs for standards", "last N submissions", or "write CODING_STANDARDS".
- ▌ Setup Matt Pocock Skills · caneff bundleConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.
- ▌ Improve Codebase Architecture · caneff bundleScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
- ▌ Thermo Nuclear Code Quality Review · caneff bundleRun an extremely strict maintainability review for abstraction quality, giant files, and spaghetti-condition growth.
- ▌ Audit Instructions · caneff bundleAudit instruction text against Anthropic's current published guidance and report what to delete.
- ▌ Docstring Coverage · caneff bundleFind public functions and classes with no docstring on a typed (py.typed) library, so consumers get docs on the surface they build against.
- ▌ Read The Damn Docs · caneffUse when installing or upgrading a version-sensitive package, SDK, or framework; working with auth, billing, migrations, or another security-sensitive flow; or diagnosing an error that suggests API drift (deprecation, missing exports, changed defaults, version mismatch). Web-searches for current official docs and reads them before assuming from memory. For Anthropic/Claude API questions, defer to the claude-api skill instead.
- ▌ Skills Safe Update · caneff bundleUpdate agent skills installed via the `npx skills` package manager without losing local edits — including following upstream renames, folder moves, and rewrites that a plain update would drop as orphans.
- ▌ Wasm Compatibility · caneff bundleCheck if a marimo notebook is compatible with WebAssembly (WASM) and report any issues.
- ▌
- ▌ Setup Peacock Color · caneff bundleSet a VS Code Peacock window color for the current repo — writes .vscode/settings.json with a base color plus all derived activity/status/title-bar customizations.
- ▌
- ▌ Setup TS Deep Modules · caneff bundleWire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files.
- ▌ Multi Axis Code Review · caneff bundleReview the changes since a fixed point (commit, branch, tag, or merge-base) — or the union of a list of landed shas — along three axes — Standards (does the code follow this repo's documented coding standards, the Fowler smell baseline, and the over-engineering lens?), Spec (does the code match what the originating issue/spec asked for?) and Correctness (how does it fail in the field, and does every new test actually witness its claim?). Runs the three reviews in parallel sub-agents, waits for all of them, and reports every finding side by side. It replaces the built-in `/code-review` in the implement lane. Use when the user wants a branch, PR, or work-in-progress diff checked against its spec, the repo's standards, and runtime failure.
- ▌ Python Testing Patterns · caneff bundlePython testing patterns and test-quality review using pytest. Use when writing, modifying, refactoring, or reviewing Python tests; before adding assertions, apply a critical-eye pass for behavior ownership, duplicate coverage, brittle string matching, overbroad fixture use, and unit-vs-integration boundaries.
- ▌ Mutation Audit · caneff bundlePoint at ONE module to learn which of its tests pass without actually catching a bug — run mutmut, scrape the survivors, emit test-audit findings. Opt-in, never in the default sweep.
- ▌ Ponytail Audit · caneff bundleWhole-repo audit for over-engineering: scans the entire codebase for a ranked list of what to delete, simplify, or replace with stdlib/native equivalents. Use when the user says "audit this codebase", "audit for over-engineering", "what can I delete from this repo", "find bloat", "ponytail-audit", or "/ponytail-audit". One-shot report, does not apply fixes.
- ▌ Type Tightness · caneffAudit loose typing the type-checker still accepts — Any where a real type is knowable, ignore-comments with no reason, and fake "boundary" excuses.
- ▌ Add Molab Badge · caneffAdd "Open in molab" badge(s) linking to marimo notebooks. Works with READMEs, docs, websites, or any markdown/HTML target.
- ▌ Auto Paper Demo · caneff bundleMake a demo of a research paper in a marimo notebook fully automatically without extra user input.
- ▌ Create Lmd Page · caneff bundleGenerate a Logic Masters Deutschland puzzle page (HTML) from a ruleset, a SudokuPad link, and an image id.
- ▌ Diagnosing Bugs · caneff bundleDiagnose hard bugs, hangs, and performance regressions. Use when the user says "diagnose" or "debug this", or reports something broken, failing, hanging, getting cut off, behaving weirdly, or stuck waiting.
- ▌ Implement Paper · caneff bundleImplement a research paper as an interactive marimo notebook together with the user. Start by understanding what the user wants to explore, fetch the paper via alphaxiv, then build a focused notebook.
- ▌
- ▌ Setup Pre Commit · caneff bundleSet up Husky pre-commit hooks with lint-staged (Prettier), type checking, and tests in the current repo.
- ▌
- ▌ Setup Python Repo · caneff bundleSet up a new Python repo — or convert an existing one — to my standard 2026 agent-assisted stack (uv, ruff, ty, pytest, pre-commit, GitHub Actions, justfile, AGENTS.md).
- ▌ Test Audit · caneff bundleRuthlessly audit a repo's tests — cut the ones that prove nothing, rewrite the ones checking the wrong thing, keep only what earns its place.
- ▌ To Tickets · caneff bundleBreak a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker — edges as text in one file per ticket locally, or native blocking links on a real tracker.
- ▌ Duplication · caneff bundleFind the same logic written twice — exact copy-paste and "same data decoded two ways" semantic duplicates — so a second home for one behavior gets caught before it drifts.
- ▌ File Ticket · caneff bundleTurn a review finding or chat conclusion into a tracked issue, one-shot. Use when the user says "file it", "file 1", "file a ticket for X", or "file tickets for X and Y" — a capture request, not a spec to slice or a batch to triage.
- ▌ Find Skills · caneffDiscover, evaluate, and install agent skills from the open skills ecosystem. Use when the user asks "what other skills", "skills like X", "which skills should we adopt", "find a skill for", or shares a skills repository to assess.
- ▌ Marimo Pair · caneff bundleDrive a live marimo notebook as a workspace: run Python in the same kernel the user does, inspect live notebook state, and commit durable notebook changes. Use when the user wants to start a marimo notebook or pair on an active marimo session.
- ▌ Skill Audit · caneff bundleAudit ~/.agents/skills for model-invocable skills that cost context but go unused, so you can flip stale ones to `disable-model-invocation: true`.
- ▌ Skills Sync · caneff bundleReport and optionally repair drift between the two skill directories (~/.agents/skills canonical bodies, ~/.claude/skills symlinks).
- ▌ Visual Plan · caneff bundleTurn ordinary text plans into rich interactive visual plans with diagrams, file maps, annotated code, open questions, and UI/prototype review when useful.
- ▌ Computer Use · caneff bundleDrive or inspect a visible Windows window from WSL — read a window's title or content, click, type, or screenshot it. Use when Chris says "look at that window", "click X in <app>", "what does <window> say", or otherwise names a specific visible window or app to inspect or control.
- ▌ Domain Drift · caneffAudit whether code names match the project's own domain vocabulary — a generic name standing in for a defined term, two names for one concept, or a term used for the wrong thing.
- ▌ Tdd · caneff bundleTest-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
- ▌
- ▌ Landed · caneff bundleRender recent commits across the workspace's repos as a local HTML review page, and open it in the browser. Use when the user asks what landed, what was committed recently, wants to review recent agent work, or asks to open the landed page.
- ▌ Pickup · caneffPick up work from the most recent handoff document left in /tmp. Use after a /clear or in a fresh session to resume where a /handoff left off.
- ▌ Marimo Batch · caneff bundleAn opintionated skill to prepare a marimo notebook to make it ready for a scheduled run.
- ▌ Visual Teach · caneff bundle`visual-teach` is the styling + behavior floor under `/teach` lessons and the audit-skill HTML reports — link its Base first, then use ready-made `vt-*` blocks for the standard parts. Use the blocks for the commodity parts and build your bespoke exercise or report content on top, never instead of it.
- ▌ Comment Audit · caneffRuthlessly audit a repo's comments — delete every one that doesn't earn its place, keep only what the code cannot say.
- ▌ Prompt Master · caneff bundleGenerates optimized prompts for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.).
- ▌ Error Handling · caneff bundleFind swallowed errors — bare excepts, `except Exception: pass`, silent drops — and sort a justified silence from an unjustified one, enforcing the repo's fail-loud rule where a plain linter stops short.
- ▌ Implement Spec · caneff bundleDrive one sliced spec as a nested run — the burn's worker, its own slices' controller, with an exploration pass, a closing ticket and a spec-level review.
- ▌ Triage · caneff bundleMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.
- ▌ Wizard · caneff bundleGenerate an interactive bash wizard that walks a human through steps only they can perform.
- ▌ Handoff · caneff bundleCompact the current conversation into a self-contained handoff document a fresh session can pick up. Default writes it to /tmp and tells you to /clear and /pickup; a leading mode word (agent / sub / side) instead launches an independent session, a sub-agent, or a background side-agent seeded with it.
- ▌ Sm Link · caneffInspect and edit SudokuMaker puzzle links (sudokumaker.app) — report a link's size, givens, ring state, entered cells, constraints and verdict, or decode a link to JSON, edit it, and encode it back. Use when asked to decode, inspect, check, edit, or rebuild a SudokuMaker link, or to check whether a link is safe to share.
- ▌ To Spec · caneff bundleTurn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.
- ▌ Ask Matt · caneff bundleAsk which skill or flow fits your situation. A router over the skills in this repo.
- ▌ Burndown · caneff bundleDrain a mixed-origin ticket queue to empty as one supervised run, frontier-first, one PR per ticket.
- ▌ Grilling · caneff bundleGrill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.
- ▌ Research · caneff bundleResearch a question from high-trust primary sources and capture the findings as Markdown in the repo. Use when the user says "do some research", "research about", "look into", or asks for options, comparisons, current capabilities, or evidence about a tool, workflow, or practice. Version-sensitive package, SDK, or API docs go to read-the-damn-docs instead.
- ▌ Dead Code · caneff bundleFind code nobody calls — dead functions, classes, unreachable branches, stray imports — and sort real dead code from code that only looks unused.
- ▌ Humanizer · caneff bundleRemove signs of AI-generated writing from text so it reads as natural, human-written prose.
- ▌
- ▌
- ▌ Wayfinder · caneff bundlePlan a huge chunk of work — more than one agent session can hold — as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.
- ▌ All Audits · caneff bundleRun every repo audit at once — thirteen audit skills in parallel, one HTML index linking each report, then grill through them one at a time and land the sweep as one spec.
- ▌ Crap Audit · caneff bundleScore every function's CRAP (Change Risk Anti-Patterns — complexity squared times uncovered fraction cubed, plus complexity) from a fresh test run and rank the real risk hotspots. Slash-only.