Claude Skills by ericmjl

Proposes updates to AGENTS.md so repo-local coding-agent instructions stay accurate, non-contradictory, and consolidated. Use when the user corrects the agent (e.g. do not do X, always do Y, from now on, remember this), asks to edit or sync AGENTS.md, CLAUDE.md, .claude/CLAUDE.md, or GEMINI.md, wants one canonical instruction file or to deduplicate agent docs, or flags conflicting or outdated repository agent rules.

Break unstaged or mixed changes into small, atomic, logical git commits with Conventional Commit messages. Use when the user asks to commit, stage, or split changes — e.g. 'help me commit', 'commit my changes', 'break these into logical commits', 'what should I commit first', 'stage my changes', 'commit atomically', or 'make atomic commits'. Uses git diff, git add -p, and Conventional Commits (feat:/fix:/refactor:/chore:/docs:/test:/style:/perf:/build:/ci).

Improve coherence in drafts by diagnosing broken flow, rewriting transitions, aligning paragraph purpose, and preserving author voice. Use when a user says things like 'make this coherent', 'tighten the flow', 'this feels choppy', 'can you smooth transitions', or asks to revise prose in Markdown, notes, essays, blog posts, emails, or docs so ideas connect cleanly from sentence to sentence and section to section.

Instinct-based learning system that observes sessions via hooks, creates atomic instincts with confidence scoring, and evolves them into skills/commands/agents. v3.0 adds project-scoped instincts to prevent cross-project contamination.

Perform comprehensive cybersecurity code review across 9 security dimensions: injection prevention, authentication/authorization, secrets management, supply chain security (including CI/CD pipeline integrity), cryptography, secure configuration/API security (including SSRF), error handling/logging/resource safety, LLM/AI application security, and infrastructure/API protocol security (GraphQL, Kubernetes, WebSockets, OAuth 2.0, gRPC). Use when reviewing newly written code, auditing existing re...

Debug GitHub Actions and CI failures using gh CLI: watch PR checks, fetch run logs, analyze failures, reproduce locally, and suggest fixes. Use when CI fails, when the user wants to watch or verify CI, or when debugging workflow runs.

Guide for design-driven development with prescribed folder structure. New features use full workflow (HLD → LLD → EARS). Bug fixes skip doc creation but verify intent coherence.

Breaks the cycle of repeated failed attempts by redirecting the agent to find the authoritative source of truth about a system's structure before trying again. Trigger this skill when: (1) the same command or query has failed 2+ times with variations, (2) output is garbled/binary and unreadable, (3) guessed field names, endpoints, or flags keep returning errors, (4) file/directory scanning is producing no useful signal. In short: if you're brute-forcing and it's not working, stop and use this...

Generate a plain-language activity report of your GitHub work for a given date range, including commits, pull requests, reviews, issues, and comments. Use this when you need to summarize what you've accomplished on GitHub for status updates, retrospectives, or tracking your work.

Use GitHub CLI (gh) for common operations like creating PRs, viewing GitHub Actions logs, managing issues, reviewing PRs, and more. When merging PRs via gh, prefer rebase merge over squash or merge commits unless repo policy or the user explicitly requests otherwise.

Reports your GitHub activity for a specific day using the gh CLI, showing commits, issues, and activity timeline. Use this when you need to review what you accomplished on GitHub for a particular date.

PR code review workflow: pull diff, two-pass findings, draft and prioritize comments, then optionally post via gh api.

Create single-file HTML slide presentations with vanilla JS/CSS. Themed (terminal.css, catppuccin, nord), keyboard-navigable, with inline SVG diagrams and animations. Use when the user asks for an HTML presentation, slide deck, or single-file slides without a framework.

Use the MemPalace CLI to persist, search, and manage AI memory across sessions. Triggers whenever the user asks to remember, recall, save, store, or retrieve information that should persist across conversations — including past decisions, preferences, facts about people or projects, conversation history, and any verbatim content the user wants preserved. Also triggers when the user mentions "palace", "wings", "rooms", "drawers", "knowledge graph", "AAAK", or when they want to search through h...

Conduct machine learning experiments from planning through evaluation and report writing. Use when running ML experiments, testing hypotheses, training models, or writing up results. Covers single-hypothesis scoping, fast iteration loops, targeted logging, JOURNAL.md protocol, data-backed diagnostic plots, and scientific report writing.

Fill static PDF forms (without fillable fields) using text search for text-based PDFs or Claude Vision for image-based/scanned PDFs. Use when: (1) User wants to fill out a PDF form, (2) PDF has no fillable form fields, (3) User asks to fill in or complete a PDF form. Auto-detects PDF type and uses appropriate method.

Quickly publish markdown notes to styled Google Docs using pandoc and a Word template. Use when creating Google Docs from markdown, applying branding to documents, or pushing content to Drive for sharing.

Defensive exploratory data analysis for scientific data (CSV, FASTA, etc.). Context-first, human-guided; one plot at a time, ask why before executing, append-only journal per session, scripts with PEP723 and uv run, WebP plots. Use when opening data files for EDA or when the user wants guided scientific data exploration.

Guide for authoring Agent Skills with strong YAML `description` triggers, progressive disclosure, and bundled resources. Use when creating or updating a skill, running init_skill.py or package_skill.py, or improving a bland skill description so agents load the skill on the right user tasks.

Use Slack CLI (`slack`) for app lifecycle and workspace operations including login/logout, create/init, run/deploy, manifest validation, trigger management, app install/uninstall, environment management, and diagnostics. Use when users ask to run or troubleshoot commands like `slack create`, `slack run`, `slack deploy`, `slack trigger`, `slack manifest`, `slack auth`, `slack activity`, `slack datastore`, `slack env`, or request help-first progressive discovery with `slack help` and `slack SUB...

Break long-lived pull request branches into a mergeable stack of small PRs with clear dependency order and story flow. Use when a branch has grown too large, when the user asks to split a PR into stacked PRs, or when each PR must be reviewable in about 5 minutes while preserving logical narrative across the stack.

Advanced negotiation and communication advisor grounded in Chris Voss's tactical empathy methodology (Never Split the Difference, The Black Swan Group). Use this skill whenever the user needs help with any interpersonal situation involving influence, persuasion, or navigating difficult dynamics. This includes but is not limited to: analyzing conversations, call transcripts, or email threads; preparing for negotiations (salary, vendor, client, partner); drafting tactful responses; handling pus...

Ingest YouTube videos into the vault. Triggers when user pastes a YouTube URL (youtube.com/watch or youtu.be). Fetches transcript using yt-dlp, extracts metadata, creates transcript note and summary note. User may provide additional context about the video.

Comprehensive YouTube operations using yt-dlp - download videos/audio, extract transcripts and subtitles, get metadata, work with playlists, download thumbnails, and inspect available formats. Use this for any YouTube content processing task.