Top Agent Skills

Connect Claude to any app. Send emails, create issues, post messages, update databases - take real actions across Gmail, Slack, GitHub, Notion, and 1000+ services.

Generate favicons from a source image

GDB debugging assistant for AI agents - analyze core dumps, debug live processes, investigate crashes and deadlocks with source code correlation

Five-phase pipeline (triage → clarify → research → PRD → plan) that turns a raw idea into four linked files: clarifying questions, deep research, a PRD with non-goals and metrics, and a phased execution plan with mermaid user journey and kill criteria.

Presentation creation, format conversion (ODP/PPTX/PDF), slide automation with LibreOffice Impress.

Algorithm-first discipline: state Big-O, data structure, and algorithm family BEFORE writing loops, queries, or recursion. Catches O(n^2), N+1, and brute-force defaults.

LLM Operations -- RAG, embeddings, vector databases, fine-tuning, prompt engineering avancado, custos de LLM, evals de qualidade e arquiteturas de IA para producao.

CRITICAL: Use for MolyKit AI chat toolkit. Triggers on: BotClient, OpenAI, SSE streaming, AI chat, molykit, PlatformSend, spawn(), ThreadToken, cross-platform async, Chat widget, Messages, PromptInput, Avatar, LLM

Onboard

Write idiomatic PHP code with generators, iterators, SPL data structures, and modern OOP features. Use PROACTIVELY for high-performance PHP applications.

Reads a public API using a declared token.

Looks harmless but omits important runtime requirements.

Review governed memory proposals, policy tags, tombstones, and effective memory.

Inspect a project and propose safe cleanup of generated or cache files.

Prepare release readiness notes, version checks, test checklist, and local issue drafts.

Review dependency manifests, lockfiles, and risky install surfaces.

Inspect local project structure, git state, docs, and relevant governed memory.

Perform read-only web research with governed browser proposals for manual review.

Create a daily business brief from local docs, memory, tasks, and safe web context.

Plan cafe marketing work across daily posts, promotions, feedback, and local competition.

Review protected asset policy and risky local data surfaces.

Use when driving a feature prompt to a production-ready PR through discovery, definition, design, spec, issues, and dev phases, or when the user runs /feature or /feature resume — orchestrates the A-Team agentic pipeline over a target repo.

Use when a feature's screens and flows are drafted and the design phase needs a navigable wireframe — a multi-screen greyscale Astro prototype where buttons and links actually work, suitable for the design gate and moderated user testing. In the A-Team pipeline it is conducted by ateam-design and consumes design.md's

prd-writer

ticket-writer

discovery-plan

product-report

Use when a project's durable context needs to be seeded or refreshed — creating or updating docs/product/context.md (overview + digest + source index + glossary + Know/Don't-Know ledger + design context + technical context) in the target repo; when a project has no context layer yet (Stage-0 setup); when domain terminology is still forming or has drifted; after a meeting, new evidence batch, or scope change that outdates the current context. Also use to ingest new docs/product/input/ batches into the digest. Do not use for repo scaffolding or CI setup, for writing the discovery plan, PRD, or epics (use discovery-plan, prd-writer, epics), or for creating or revising Jobs to be Done — this skill records finished JTBDs by id and marks missing ones TBD; jobs-to-be-done owns minting them.

ateam-definition

research-synthesis

Load this skill when the user says "close issue", "close-issue", or "/close-issue". Verifies IT and UC acceptance criteria for a story issue, ticks off passing checkboxes, and closes the issue only if all criteria pass and the linked PR has been merged.

Load this skill when the user says "prepare base", "prepare-base", or "/prepare-base". Prunes remote tracking references, checks out a base branch, pulls latest, and cleans up local-only feature branches after confirmation.

Load this skill when the user says "review issue", "review-issue", or "/review-issue". Reviews an implemented GitHub issue against its requirements and the project coding standards, acting as a critical tech-lead review pass before merging.

Load this skill when the user says "resolve comments", "resolve-comments", or "/resolve-comments". Checks open PR review comments, addresses valid ones in a single batched commit, and resolves all threads with appropriate replies.

Load this skill when the user says "asvs security review", "asvs-security-review", "/asvs-security-review", or when invoked by reference from the review-issue skill's standards-review step. Performs a rule-by-rule walk of docs/security-standards.md against a diff, scoped to the ASVS sections the diff actually touches, and reports findings using the project's existing severity mapping.

Load this skill when the user says "update coding standards", "update-coding-standards", or "/update-coding-standards". Analyzes the diff of a merged PR and appends any valuable new patterns to the relevant coding standards doc — but only if something genuinely warrants documenting.

Explain any concept in plain words with one concrete analogy - clear enough for a smart child or a busy outsider, never condescending, never wrong.

Compress long text into a one-line takeaway plus the few points that matter - numbers and meaning-changing caveats preserved, nothing invented.

Diagnose and fix a bug through a disciplined reproduce, isolate, diagnose, fix, verify loop driven by evidence and one hypothesis at a time — not guesswork.

Produce a daily or weekly cross-source digest — mentions, action items, and decisions that need attention, not raw activity.

Build a personal budget from income and expenses using the 50/30/20 rule or a zero-based plan.

Structure a decision into steelmanned, impact-weighted pros and cons - irreversible consequences flagged, the decisive unknown named, and a clear recommendation you can reject.

Produce the weekly regulatory digest — a prioritized rollup of new rules, open comment deadlines, and compliance gaps across every tracked matter.

Write the private-equity investment committee memo for a deal — thesis, company and market, financials and QoE, valuation and returns with base/bull/bear MOIC and IRR, value-creation plan, risks and mitigants, structure and financing, sensitivities, and a conditioned recommendation.

Front door to the small-business toolkit — routes a vague or specific request to the best skill and explains what's available.

Research a customer question across docs, changelogs, and past tickets, then answer with source attribution, a confidence rating, and honest gaps.

Turn a finished design into a developer-ready handoff spec - tokens, every interactive state, responsive breakpoints, and the edge cases mockups never show.

Disable an installed legal skill without uninstalling it — make it inert and unfireable while preserving its version, config, and provenance.

Produce a decision-ready legal briefing — daily digest, topic research, or incident write-up — scoped to what a non-lawyer stakeholder needs to decide or act on.

Write QML unit tests with Qt Quick Test using TestCase, SignalSpy, and data-driven functions.

Onboard a new small-business owner — connect their first two tools, run one recipe, interview them, store context, and set a weekly check-in.

Optimize a LinkedIn profile — headline, about, experience, and keywords — for recruiter search and credibility.

Build a character with a clear want, need, flaw, distinct voice, and arc.

Run a developmental then line-edit pass on a fiction draft, big issues before small.

Answer any data question, scaling effort from a one-number lookup to a full stakeholder-ready analysis with a recommendation.

Rewrite a blunt or rough email draft into a warm, friendly-but-professional version - tone and clarity improved, intent, facts, names, dates, and requests preserved exactly.

Prepare a DMCA 512(c) takedown notice or a 512(g) counter-notice for online copyright infringement with all statutory elements.

Design a Qt Quick (QML) user interface with sound layouts, theming, responsiveness, and accessibility.

Assemble tax-season materials — quarterly estimates or year-end 1099s — plus an organized accountant packet.

Add safety guardrails to an LLM app — prompt-injection defense, output validation, PII handling, and a clear refusal policy.