dotfiles

This skill should be used when the user asks to "commit and push", "commit push", "sync changes", "push changes", "commit and sync", or "update remote". Handles the full workflow of committing changes, pulling with rebase, and pushing to remote.

Sweep recent AI safety research from curated sources (Anthropic alignment science / red team, OpenAI, GDM, Apollo, Redwood, METR, FAR AI, Truthful AI, alphaxiv, arXiv) and surface items matching tracked topic terms (inoculation prompting, reward hacking, exploration hacking, metagaming, eval gaming, OOCR, scheming, alignment faking, sandbagging, etc.). Use when asked to "sweep AI safety", "what's new in alignment", "any recent papers on X", "weekly safety digest", or for staying current on AI safety literature.

Anthropic visual style for plots, diagrams, slides, and web. Use when creating any visual output that should have Anthropic's look-and-feel — matplotlib charts, TikZ diagrams, HTML/CSS, or presentations.

Catch LLM-fabricated citations in BibTeX files. Verifies arXiv/OpenReview entries against live metadata (titles, first authors), then guides manual verification of authorless prose claims. Use before submitting papers, after any LLM-assisted citation generation, or when a reference smells off.

Fact-check prose claims in slides, reports, PDFs, and papers — statistics, comparatives, attributions, causal claims, quotes. Two-pass extract-then-verify protocol with strict numerical precision and a doc-only mode. Use when the user asks to "check the claims in this deck", "fact-check this report", "audit this PDF", "verify the numbers in these slides", or before publishing/shipping any externally-facing document with quantitative claims. Complements `check-bib-references` (which handles BibTeX entries) — this skill handles the prose around them.

Log a one-line knowledge gap to the project's gaps.md file. Use when the user is surprised by Claude's answer, says "I didn't know that", "wait what", or wants to record a misconception they just discovered. Format "I assumed X but actually Y". Personal misconception log — much higher learning signal than feedback memories.

Resurface a random sample of feedback memories for spaced-repetition review — "still true? changed? promote to global rule?". Use when user asks for feedback retrospective, weekly memory review, or to audit accumulated coaching corrections. Also good for periodic via /schedule.

Set up project-local messaging channels (Telegram, iMessage, Things Cloud). Platform-aware — Telegram works everywhere, iMessage requires macOS. Extensible dispatch table — adding a new channel means adding one row.

Move a repo to a new directory. Handles venv, Claude Code project state, path references, and tmux sessions.

Submit experiments or agent jobs with resource limits, check queue status, pause/resume workloads, troubleshoot slow machine

Run Python code in the cloud with serverless containers, GPUs, and autoscaling. Use when deploying ML models, running batch processing jobs, scheduling compute-intensive tasks, or serving APIs that require GPU acceleration or dynamic scaling.

Merge current worktree branch into the original branch, resolve conflicts with AI, then mark worktree for cleanup

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup

Commit current work with optional message. Handles git status/diff checking and message formatting.

Orchestrate agent teams for parallelizable multi-faceted tasks. Use when work requires inter-agent communication, competing hypotheses, or multi-file parallel implementation.

Compact current work with optional focus using agent-context-summariser

Externalise important parts of the current conversation into an .md file for handover.

Run custom usage analytics on all Claude Code sessions