claude-brainlab

In the author's own words, from each skill's SKILL.md — showing the first 30 of 67 files.

• academic-plotting

  Generates publication-quality figures for ML papers from research context. Given a paper section or description, extracts system components and relationships to generate architecture diagrams via Gemini. Given experiment results or data, auto-selects chart type and generates data-driven figures via matplotlib/seaborn. Use when creating any figure for a conference paper.

  skills/academic-plotting/SKILL.md on GitHub (opens in a new tab)

• agent-identifier

  Use when creating or configuring Claude Code agents and their frontmatter.

  skills/agent-identifier/SKILL.md on GitHub (opens in a new tab)

• architecture-design

  Use only when creating new registrable ML components that require Factory or Registry patterns.

  skills/architecture-design/SKILL.md on GitHub (opens in a new tab)

• astar-paper-review

  This skill should be used when the user asks to review, referee, or critique an ML/AI paper for an A* venue or laboratory pre-review. It runs a grounded multi-agent audit of theory, prior work, experiments, and reproducibility.

  skills/astar-paper-review/SKILL.md on GitHub (opens in a new tab)

• bug-detective

  Investigates software failures, exceptions, and errors by tracing execution, reproducing the problem, and testing the suspected cause. Use when the user reports a crash, a traceback, an unexpected result, says 'this code throws an error', 'why does this fail', or asks to debug a failing function.

  skills/bug-detective/SKILL.md on GitHub (opens in a new tab)

• call-notes

  Turns a research call transcript into a merge request for the lab work it was about: the call note, edits to the claims that were discussed, and tasks for the work's GitLab board without duplicates. Use when the user says 'запиши звонок', 'call notes', 'разбери созвон', 'разбери митинг', or pastes a transcript or meeting summary to be filed. Also the contract for Brain Call's headless analysis.

  skills/call-notes/SKILL.md on GitHub (opens in a new tab)

• caveman

  Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.

  skills/caveman/SKILL.md on GitHub (opens in a new tab)

• citation-verification

  This skill provides reference guidance for citation verification in academic writing. Use when the user asks about "citation verification best practices", "how to verify references", "preventing fake citations", or needs guidance on citation accuracy. This skill supports ml-paper-writing by providing detailed verification principles and common error patterns.

  skills/citation-verification/SKILL.md on GitHub (opens in a new tab)

• code-ingest

  Adds a code repository to the Obsidian code library with its purpose, usage, and links to research projects. Use when the user hands over a repo URL or path and says 'add to the code library', 'разбери репозиторий', or starts a project that uses a repo missing from Code/.

  skills/code-ingest/SKILL.md on GitHub (opens in a new tab)

• code-library

  Finds documented repositories, execution patterns, and project links in the user's Obsidian code library. Use when the user asks 'how does repo X work', 'where is the code for this project', how to run or monitor experiments, or which code skills exist.

  skills/code-library/SKILL.md on GitHub (opens in a new tab)

• code-review-excellence

  This skill should be used when the user asks to review a diff or pull request, write review comments, audit code quality, establish review standards, or improve how a team performs code review.

  skills/code-review-excellence/SKILL.md on GitHub (opens in a new tab)

• command-development

  Creates and repairs Claude Code slash commands, including their frontmatter, arguments, prompts, and local file references. Use when the user says 'create a slash command', 'add a command', 'fix my command arguments', or edits a markdown file under a commands directory.

  skills/command-development/SKILL.md on GitHub (opens in a new tab)

• create-project

  Creates a new lab work (one GitLab repository in its theme, cloned into ~/brainlab at the same path), or repairs an existing work's setup, after checking that no such work already exists. Use when the user says 'create a project', 'new project', 'заведи проект', 'заведи работу', 'set up a repo for a paper', or a work is missing its README, people lines, local folders or clone settings.

  skills/create-project/SKILL.md on GitHub (opens in a new tab)

• daily-coding

  Use for everyday coding tasks that involve writing or modifying source code.

  skills/daily-coding/SKILL.md on GitHub (opens in a new tab)

• defuddle

  Extract readable Markdown from a web page with Defuddle when page navigation obscures the content.

  skills/defuddle/SKILL.md on GitHub (opens in a new tab)

• diagnose

  Diagnoses hard bugs and performance regressions by building a reproducible feedback loop and testing competing causal hypotheses. Use when a bug is intermittent or resists a quick fix, the user says 'it is flaky', 'performance regressed', 'I cannot reproduce it', or an earlier fix attempt failed.

  skills/diagnose/SKILL.md on GitHub (opens in a new tab)

• doc-coauthoring

  Co-authors a technical document, proposal, or specification through context gathering, targeted drafting, and reader-testing review. Use when the user says 'help me write a doc', 'draft a spec', 'write a proposal', 'RFC', or wants to iterate on a design document together.

  skills/doc-coauthoring/SKILL.md on GitHub (opens in a new tab)

• experiment-runs

  Запуск обучения на серверах лаборатории и воспроизводимость прогона. Use when starting, monitoring or resuming a training run on a remote GPU box, choosing which server or GPU index to take, syncing code to a server, or deciding what a run must record to be reproducible later.

  skills/experiment-runs/SKILL.md on GitHub (opens in a new tab)

• frontend-design

  Designs and implements polished, production-grade frontend interfaces with a coherent visual direction and real working code. Use when the user asks to build a component, page, landing page, dashboard or app UI, or wants a frontend that looks distinctive rather than generic.

  skills/frontend-design/SKILL.md on GitHub (opens in a new tab)

• git-guardrails-claude-code

  Install Claude Code hooks that block selected destructive Git operations when the user requests guardrails.

  skills/git-guardrails-claude-code/SKILL.md on GitHub (opens in a new tab)

• git-workflow

  Prepares commits, branches and pull requests following the repository's existing Git conventions, including Conventional Commits messages and merge strategy. Use when the user asks to write a commit message, name or create a branch, open a pull request, or asks how to handle merges in this project.

  skills/git-workflow/SKILL.md on GitHub (opens in a new tab)

• google-workspace-mcp

  Reads and edits Google Docs and Drive files through the user-scope google-workspace MCP and repairs its authentication or connection. Use when the user asks to open or change a Google Doc, or when google-workspace tools report auth errors, missing tools or a failed connection.

  skills/google-workspace-mcp/SKILL.md on GitHub (opens in a new tab)

• grill-me

  Interviews the user relentlessly about a plan or design, walking each branch of the decision tree and proposing a recommended answer to every question. Use when the user says grill me, stress-test my plan, or wants to resolve assumptions and open decisions before building.

  skills/grill-me/SKILL.md on GitHub (opens in a new tab)

• grill-with-docs

  Challenges a proposed plan or design against existing domain documents, terminology and implementation choices by asking one question at a time with recommended answers. Use when the user wants a design grilled against the docs or codebase, or says check this plan against our domain model.

  skills/grill-with-docs/SKILL.md on GitHub (opens in a new tab)

• handoff-to-jarvis

  Передать запущенные прогоны серверному агенту jarvis4shkodnik под мониторинг — он подтвердит в Telegram, сверит с сервером и поднимет вахтёра. Use when the user says "передай Джарвису", "поставь под мониторинг", "отдай на сервер", "hand off to jarvis", or after launching runs he intends to leave unattended.

  skills/handoff-to-jarvis/SKILL.md on GitHub (opens in a new tab)

• handoff

  Summarizes the current conversation into a handoff document saved to a temporary file so a fresh agent can continue the work. Use when the user says hand this off, write a handoff, wrap up for the next session, or wants to continue in a new context window.

  skills/handoff/SKILL.md on GitHub (opens in a new tab)

• hook-development

  Builds and debugs Claude Code lifecycle hooks, covering event types, input payloads, JSON responses, exit codes and registration in settings or plugins. Use when the user asks to create a PreToolUse or PostToolUse hook, block or validate tool calls, or fix a hook that does not fire.

  skills/hook-development/SKILL.md on GitHub (opens in a new tab)

• improve-codebase-architecture

  Finds architectural friction in an existing codebase and proposes deepening refactors that turn shallow modules into deep ones, improving testability and navigability. Use when the user asks to improve architecture, find refactoring opportunities, consolidate tightly coupled modules, or make the code easier to test.

  skills/improve-codebase-architecture/SKILL.md on GitHub (opens in a new tab)

• json-canvas

  Creates and edits Obsidian Canvas (.canvas) files with nodes, edges, groups, and layout following JSON Canvas Spec 1.0. Use when the user asks to make a canvas, add or connect nodes, group cards, rearrange a canvas layout, or fix a broken .canvas file.

  skills/json-canvas/SKILL.md on GitHub (opens in a new tab)

• kaggle-learner

  Extracts reusable ML techniques, code patterns, and feature engineering ideas from Kaggle competition winning solutions and maps where they apply. Use when the user shares a Kaggle competition, asks to learn from top solutions, or wants competition tricks for their own model or pipeline.

  skills/kaggle-learner/SKILL.md on GitHub (opens in a new tab)

No skill matches that filter.

Similar skills

• ponytailby DietrichGebert Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. 181k stars · 6 skills
• awesome-claude-skillsby ComposioHQ A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows 78.4k stars
• marketingskillsby coreyhaines31 Marketing skills for Claude Code and AI agents. CRO, copywriting, SEO, analytics, and growth engineering. 57.5k stars · 30 skills
• khazix-skillsby KKKKhazix 数字生命卡兹克开源的 AI Skills 合集 | Agent Skills: leader（帮你定义目标）, neat-freak 洁癖, hv-analysis, khazix-writer & more — Claude Code, Codex & 40+ agents 21.8k stars · 6 skills
• Wazaby tw93 🥷 Engineering habits you already know, turned into skills AI assistants like Claude Code, Codex and Cursor can run. 7.5k stars · 8 skills
• appllama-skillsby Appllama A builder, not just a researcher. Agent skills that turn top-grossing app patterns into native-quality mobile screens. 3.9k stars · 2 skills