Skills
101,323 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Remote control of DOCA Flow over gRPC with the shipped `doca_flow_grpc` server: program Flow pipes and entries from Python, Go, Rust or Java, or from another host, instead of linking `libdoca_flow.so`. PLAINTEXT-ONLY: the server uses `grpc::InsecureServerCredentials()` with NO TLS / mTLS / token-auth knob; transport security must come from external infrastructure (e.g. an mTLS proxy / sidecar) on a trusted segment. Trigger even without "gRPC": "program Flow rules from Python on another host", "remotely configure pipes on the BlueField", "client times out connecting to the Flow server", "where is the .proto for Flow", "UNAUTHENTICATED / FAILED_PRECONDITION on a Flow RPC". Covers bringing up, configuring, hardening and debugging the server. Route elsewhere for the underlying doca-flow C API, generic gRPC tooling (protoc, language bindings), or DOCA install / BFB bring-up.
Use this skill when the user is doing hands-on DOCA Management programming against BlueField / ConnectX devices — standing up a management or representor context (doca_mgmt_dev_ctx / doca_mgmt_dev_rep_ctx), querying device caps (data-direct, caps-general), toggling congestion-control global status, modifying diagnostics-data, setting ICM quotas, or issuing a raw firmware command via doca_mgmt_raw_cmd with the right scope (CONFIGURATION / DEBUG_READ_ONLY / DEBUG_WRITE / DEBUG_WRITE_FULL). Trigger even when the user does not say "DOCA Management" — typical implicit phrasings include "fleet tool that walks every BlueField and reads device state", "toggle data-direct on a VF", "set an ICM quota per representor", "send a raw firmware command from C", "DOCA_ERROR_IO_FAILED from raw_cmd", or "fwctl ioctl is failing". Refuse and route elsewhere for mlxconfig direct operation, BFB / firmware reflash, streaming telemetry, doca_caps CLI snapshots, or DOCA install itself — those belong to other skills.
Use this skill when the user is invoking the DOCA PCC Counters tool — the `pcc_counters.sh` bash script under the DOCA tools directory — to arm and read the fixed firmware/hardware PCC (Programmable Congestion Control) diagnostic counters (CNP, RTT, WRED-drop, etc.) on a ConnectX / BlueField device via mst + the mlx5 debugfs `diag_cnt` interface. The script takes two positional args — `set | query` and an mst device path — with no `--help` or subcommands. Trigger even without "pcc_counters.sh" or "PCC counters": "how do I read the CNP / RTT / WRED-drop counters", "PCC counter stuck at zero", "the script says Bad Device", or "is congestion control dropping packets on this port?". Route elsewhere for writing a custom PCC algorithm (doca-pcc), factory firmware PCC config, DOCA install, or fleet-wide CC tuning.
Use this skill to read DOCA hardware-counter events from a `doca_dev` through the per-domain Telemetry reader libraries: `doca_telemetry_pcc`, `_dpa`, `_diag`, `_adp_retx`, `_phy`, and `_pci`. It covers capability checks, context creation, startup, and per-domain reads or samples. Trigger for implicit requests such as "read PCC counters from my BlueField app", "sample DPA counter exports", or "expose PHY, PCI, or DIAG counters from this doca_dev". This is the counter-reader surface, not a NetFlow, IPFIX, or local-socket collector. Route publishing and export to `doca-telemetry-exporter`; route the deployed DOCA Telemetry Service to `doca-dts`, and collectors and plain stdout logging elsewhere.
Raw verbs inside DOCA Core (doca-verbs): direct control of QP, CQ, PD, SRQ and address-handle objects through doca_verbs_context and doca_verbs_qp_create / doca_verbs_qp_modify / doca_verbs_cq_create, porting libibverbs code into DOCA Core, or a QP attribute, opcode or WR flag the doca-rdma task API does not expose. Trigger on “raw QP attribute the task API does not expose”, “porting libibverbs code”, “IO_FAILED on WR submit”, “QP state transition rejected”, or “attach a congestion-control group”, even without naming doca-verbs. Refuse and route elsewhere for ordinary doca-rdma / doca-eth / doca-rmax workloads, DOCA install, Core internals, and general libibverbs theory
Transform documentation into a powerful marketing channel that attracts, converts, and retains developers. This skill covers creating documentation that ranks in search, converts visitors into users, and accelerates adoption through exceptional information architecture and developer-focused writing. Trigger phrases: "documentation strategy", "docs that convert", "quickstart optimization", "API reference best practices", "developer documentation", "docs SEO", "information architecture for docs", "technical writing for marketing", "docs-led growth"
Orchestrate the full workflow for analyzing plot points and dramatic functions in long texts: preprocessing, segmented analysis, and result integration. Use when a novel or script needs a structured dramatic analysis report, or when a dramatic workflow or plot-point analysis pipeline is mentioned.
Detect infrastructure drift between Terraform state and actual cloud resources. Identifies unmanaged resources, manual changes, and configuration drift. Use when: - User asks to check for infrastructure drift - User wants to find unmanaged cloud resources - User mentions "drift detection" or "Terraform drift" - User asks to compare cloud state to IaC - User wants to audit infrastructure changes
Find evidence-backed simplifications in DeepSeek Harness code, APIs, configuration, tests, and prose; write or consolidate proposals, identify small inline cleanups, or assess simplifications from another branch. Use for removing dead, duplicated, speculative, or unnecessarily maintained behavior and infrastructure.
Use for ElevenLabs provider-specific dubbing, localization, voice changer, speech-to-speech, and voice isolation/enhancement workflows. Applies when an agent must prepare source media, choose ElevenLabs dubbing versus voice conversion, manage language/speaker/timing decisions, handle voice identity and consent, call or plan against documented ElevenLabs APIs, export dubs/transcripts, estimate cost/limits, or quality-control localized speech while avoiding generic text-to-speech guidance.
Assess whether a specific Lightning web component can serve its user's purpose as an HXL declarative widget. Use for LWC-to-widget feasibility, portability audits, and candidate screening; use experience-lwc-widget-migrate when the user wants the widget built
Build and deploy Apache Flink user-defined functions (UDFs) in Java for stream processing over Kafka. Use this skill when users want to create scalar UDFs, user-defined table functions (UDTFs), or process table functions (PTFs) in Java, deploy them to Confluent Cloud or local Docker environments, and invoke them from Flink SQL or the Table API. Trigger on: Flink UDF, custom Flink function, process table function, PTF, UDTF, Flink user defined, extend Flink SQL, stateful stream processing with Flink. Do NOT trigger for: Kafka Streams UDFs (use kafka-streams-programming skill), general Flink job development without custom functions, CDC streaming data piplines that include Flink (prefer the confluent-cloud-cdc-tableflow skill), Flink connector setup, or Kafka producer/consumer code.
Comprehensive Flutter development patterns covering widgets, testing, performance, security, and animations. Use when you need quick reference for Flutter best practices, common UI patterns, performance optimization techniques, security guidelines, or animation implementations.
Use when building a finished product ad from FLUX 3 - shot design, voiceover, action-to-word sync, evidence-gated copy, deterministic assembly, and QC gates that catch clipped audio, floating products, off-model plates, and reports that claim a pass the build did not give.
Deprecated compatibility alias for structure-codebase. Activate only for the exact legacy name folder-structure or an explicit $folder-structure invocation. Contains no independent architecture guidance and must delegate to the canonical sibling skill.
Verified 2026 implementation cookbook for Awwwards-grade front ends (474 recipes, type-checked and browser-audited). Covers motion and easing tokens, GSAP ScrollTrigger + Lenis choreography (pin, scrub, horizontal, stacked cards, SplitText), CSS scroll-driven animations, modern CSS (@starting-style, anchor positioning, @property, container queries), text effects and font pairing, hover/cursor/micro-interactions, view and route transitions, preloaders and menus, CSS/SVG/canvas backgrounds, WebGL shaders and Three.js/R3F, 21st.dev / Aceternity / Magic UI / React Bits components, OKLCH color, layout and hero composition, 30 aesthetic directions, performance and accessibility. Use when building, styling, animating or redesigning any website, landing page, portfolio or UI component, or when the user asks for an effect ("make it feel premium", "scroll animations", "smooth scroll", "parallax", "text reveal", "shader background", "page transition", "custom cursor", "bento", "marquee"), even if no library is named.
Use when Codex needs to design, build, polish, or substantially redesign frontend UI: landing pages, websites, dashboards, web apps, components, prototypes, games, or visual QA for local interfaces where distinctive production-grade design matters.
Use when generating or editing images from Flutter/Dart with Firebase AI Logic and a Gemini image model (Nano Banana), making the first call work, choosing Gemini Developer API vs Vertex AI, hitting quota, billing or App Check failures, getting empty or image-only responses, sending a user photo as input, controlling aspect ratio or size, writing the image prompt, or deciding what to test.
The strategy layer for making short-form video actually go viral — pick the goal, manufacture the emotion that drives it, then build the hook + funnel around it. Use when the user says "go viral", "make it go viral", "why isn't my content getting views", "how do I get more reach / saves / shares / followers", "my reels flop", "viral strategy", "what makes a reel go viral", or "how do I grow on Instagram / TikTok / Shorts". This is the WHY/strategy above the tactical skills (viral-hook-writer writes the hook; this decides what to make it about and why).
Agent-callable Google Calendar tools — create, update, move, search, and delete events, manage calendars, check free/busy availability, resolve event colors, and manage calendar sharing. Use when the user mentions Google Calendar or wants to schedule, find, reschedule, or share events and calendars — including requests that do not name Google Calendar explicitly, e.g. "put a meeting on my calendar Tuesday 3pm" or "am I free Friday afternoon".
One-click full-pipeline PPT: first use GordenImagePPTGen to generate an image-format PPT, then use GordenImage2PPTX to reverse-convert it into an editable .pptx. Packages and orchestrates these two sub-skills in sequence, delivering the image-based PPT + editable pptx + all intermediate artefacts together. Orchestrates image-PPT generation and image→editable-PPTX conversion end to end. Use when the user only provides a topic/content and wants a complete, both good-looking and editable PPT deliverable, or does not name a specific function.
P4 phase skill for Headless Ghidra selected function substitution. Use after P3 when an approved batch of functions should have metadata applied, be decompiled through Ghidra, and be recorded as per-function capture/substitution YAML.
Produce HeyGen avatar videos with Direct Video, Video Agent, Digital Twin, Avatar Realtime, and related avatar/voice/asset APIs. Use for provider routing, avatar and voice selection, script-to-video workflows, lip-sync from audio, backgrounds/scenes, callbacks and polling, consent and rights checks, localization handoffs, artifact custody, QA, and troubleshooting HeyGen avatar-video productions.
Create, update, validate, and submit Homebrew Casks (macOS and Linux/AppImage). Use when the user mentions Homebrew cask/cask, Homebrew/homebrew-cask, adding a new cask, updating or bumping a cask, cask token naming, generate-cask-token, sha256, livecheck, zap/uninstall, AppImage/app_image, on_linux/on_macos, cross-platform cask, or when asked to run brew style, brew audit or brew lgtm for a cask.
Design and verify auditable human oversight, approval gates, escalation paths, and safe state transitions for AI agent workflows. Use when deciding which agent actions require review, adding approve/reject or dual-control flows, preventing unauthorized autonomous effects, creating decision records, reducing rubber-stamping, or recovering safely from rejected, expired, or failed actions.
Remove signs of AI-generated writing from text. Use when editing or reviewing text to sound natural and human-written. Based on Wikipedia's 'Signs of AI writing' — catches em dash overuse, rule of three, AI vocabulary, vague attributions, inflated symbolism, negative parallelisms, passive voice. Not for agent-facing text.
Meta XR Interaction SDK (com.meta.xr.sdk.interaction.ovr) in Unity. Use when adding ANY interaction to a Meta VR or Horizon OS scene - grab (hand, distance, ray, touch), poke, teleport/locomotion, gaze, interactable UI canvases - when making an object grabbable, adding interactors to hands or controllers, creating an interaction rig, or baking interaction into a prefab or a scene whose rig loads elsewhere (dynamically, DontDestroyOnLoad, additively). Use the OVR backend unless the user explicitly asks for cross-platform. Covers QuickActionsAPI and OVRQuickActionsAPI as the default way to wire interactions, running them via unity-cli, an Editor script, or MCP, installation, and the undocumented behavior that source-reading or a Play-mode run is otherwise needed to discover.
Precisely separate the foreground subject from the background in an image. Only five subject types are supported — people, pets, products, icons, and seals — and the output is a transparent-background PNG. Trigger when the user explicitly asks for cutout, background removal, transparent background, or extraction of the above subjects; other subject types and white-background isolation are not supported.
Use this skill to plan, direct, finish, and QA provider-independent mono or stereo 180-degree, 360-degree, spherical, spatial, wide-FOV, and immersive video. It covers format choice, capture rigs, stitching, nadir and seam management, stereoscopic comfort, attention direction, stabilization, ambisonic and spatial audio, titles, captions, accessibility, edit, color, VFX, projection metadata, Vision Pro, YouTube, headset delivery, device QA, and privacy, location, and likeness rights. Do not use it for interactive XR apps, volumetric/world-model generation, game-engine experiences, or provider-specific video generation prompts.
Clone ChatGPT saved memory into Letta, then optionally enrich it with broader conversation history. Designed for a slick onboarding flow that extracts hidden saved-memory/context blocks, builds Letta-ready previews, and only asks questions at meaningful checkpoints.
Plan and create Instagram content, improve profiles, diagnose distribution from account evidence, and design scoped growth or monetization experiments. Use for Instagram-specific content and strategy; not video editing, API uploads, or other platforms.
From a scientific question of interest, get recommendations on preparing and formatting the documentation related to ICHM15 planning phase, which covers ICH M15 key assessments elements, technical criteria evaluation of model and model outcomes, appropriateness of Proposed MIDD and a Model Analysis Planning (MAP) document. Optionally upload the recommended documentation in Jinkō.
AI-powered JavaScript reverse engineering tool. Senior JavaScript reverse engineering expert assistant. Actions: collect, search, deobfuscate, understand, summarize, detect-crypto, browser, debugger, breakpoint, debug-step, debug-eval, debug-vars, script, hook, stealth, dom, page. Capabilities: obfuscated code analysis, VM cracking, Webpack unpacking, AST transformation, Puppeteer/CDP automation, anti-detection, fingerprint spoofing, crypto identification, parameter extraction, algorithm recovery, Canvas/WebGL fingerprinting, WebDriver hiding, CDP debugging, breakpoint analysis, dynamic tracing, hook injection, DOM inspection, page control.
Read this BEFORE launching any subagent (Task tool, background agents, parallel agents, best-of-N, delegating work to another agent). Consensus principles for using subagents well. Triggers: launch a subagent, spawn agents, run agents in parallel, delegate to a subagent.
Reviews, audits, and modifies CockroachDB cluster settings. Self-Hosted has full control over all settings and start flags. Advanced/BYOC can modify most SQL-level settings but infrastructure settings are managed by CRL. Standard has limited settings access — session variables are the primary tuning mechanism. Basic has minimal settings — use session variables and Cloud Console. Use when auditing configuration, tuning performance, or troubleshooting settings-related issues.
Validate markdown files against formatting standards. Auto-detects and installs markdownlint-cli if missing. Checks project config or falls back to sensible defaults. Supports auto-fix mode. Use when creating or modifying markdown files, validating documentation, or enforcing markdown conventions.
Use this skill when users need to calculate market size (TAM/SAM/SOM), assess market opportunity, validate market potential, or determine if a market is big enough to pursue. Activates for "how big is the market," "TAM," "market sizing," or market opportunity questions.
Run OpenMMDL molecular dynamics workflows via the FastFold Workflows API (`openmmdl_v1`) from local topology + optional ligand files, prepare draft scripts, execute drafts, wait for completion, fetch artifacts/metrics, and extract trajectory frames. Use when users ask for OpenMMDL, protein-ligand MD, OpenMMDL script preparation, or `/openmmdl/results/<workflow_id>` reruns.
Analyze official Megatron-LM commits, PRs, and branch change sets to identify feature evolution, candidate breaking changes, and migration-relevant events. Use when Codex already has a normalized Megatron change set and needs to explain what changed, which new features matter, and which changes should flow into MindSpeed adaptation work.
Track and normalize change requests against the official Megatron-LM repository by branch, PR, commit, commit range, or time window. Use when Codex needs to collect the exact upstream change set before deeper analysis, especially for branch-aware Megatron and MindSpeed migration work, daily/periodic tracking, or preparing inputs for change analysis and migration generation.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.
