Explore

Use when deciding between ue-mcp's native category actions and Epic's wrapped ToolsetRegistry tools (the `epic_*` actions, incl. the Blueprint graph DSL) for a task in Unreal. Pulls in when authoring Blueprint graph bodies, or any time both a native action and an epic_* action could do the job and you need to pick.

Use when authoring or modifying Unreal Blueprint assets through ue-mcp. Covers the read-then-write discipline, node/pin wiring, component (SCS) hierarchy, variables, interfaces, compilation, and the difference between pin defaults and linked inputs. Pulls in any time the user asks to create, edit, or inspect a Blueprint.

Use when creating, modifying, retargeting, rigging, constraining, or validating skeletal animation through UE-MCP. Covers native UE 5.8 IK Rig and Retargeter authoring, the Control Rig begin/read/apply/bake loop, generic contact locks, per-rig anatomical and mirrored-axis discovery, quaternion keying, deterministic bone analysis, and exact fixed-frame Unreal capture.

Install and self-host deco Studio (the open-source control plane) on the user's own infrastructure. Use when the user wants to install, run, self-host, or deploy Studio locally (Docker, Rancher Desktop, kind, minikube) or on Kubernetes (managed or self-managed), or asks to "install Studio", "self-host decocms", configure the Studio Helm chart, or troubleshoot a Studio install. Detects the environment, picks the right tier, configures the charts correctly, validates end-to-end, and fixes the common failure modes.

Build or style React UI with the decocms product design system (@decocms/ui). Use when creating interfaces, pages, or components in a project that should look like decocms products, when the user mentions "design system", "@decocms/ui", "decocms style", or asks to make UI consistent with Studio. Covers installation, Tailwind v4 token setup, component inventory, and usage conventions.

Read and extract cell data from .xlsx Excel spreadsheets as TSV. Use when the user uploads or references an Excel file to read or summarize.

Script against the org's MCP tools from inside the sandbox — discover schemas under .deco/tools/, call one tool from the shell with `typegen call`, or batch many calls in a script. Use whenever a task needs repeated or bulk tool calls (same tool many times, one call per file/row/item) instead of one-at-a-time agent tool calls.

Render any text file (HTML, markdown, config, SQL, email) from a mustache template plus JSON data. Use when an output's shape is fixed ahead of time and only the values change.

Interview the user about how their team works, then set each task board column's instructions, skills and "run the agent" rule. Inserted by the board's Agentic Setup button.

Create and edit presentation decks as single self-contained HTML files with a live, editable preview and print-to-PDF export. Use when the user wants slides, a deck, or a presentation.

Screenshot a page (deploy preview or a dev server on localhost) in the sandbox's headless Chromium, look at it, and embed it in a task comment. Use to verify a visual change or record before/after evidence.

Read and inspect existing .pptx PowerPoint files (text, speaker notes, slide images). Use only for reading uploaded PowerPoint files — to AUTHOR a deck use the slides skill instead.

Read and extract text from .pdf files, page by page. Use when the user uploads or references a PDF to read or summarize.

Check that the fix a Jira issue carries is live on the PRODUCTION site and behaves as the issue asked, then close the card, or open a follow-up issue for what failed. Insert into the prompt of a run started on a batch of merged issues, or of a Jira post-deploy column rule.

Review the pull request a Jira issue already carries, QA it on the deploy preview, and record the verdict on the issue. Insert into the prompt of a Jira review or QA column rule.

Assemble a release from a batch of Jira issues — one release branch, pull request, preview and tracking issue per repository — and, when a person says to ship it, merge it and move the cards on. Insert into the prompt of a run started on a batch of issues.

Decide whether a reviewed Jira issue can skip the team's own QA and go straight to the next column, or stays for a person. Insert into the prompt of the Jira column reviewed work waits in before the client sees it.

Land the pull requests a set of Jira issues carry, in order, resolving a conflict when one appears, and move each card on. Insert into the prompt of a run started on a batch of issues, or of a Jira column rule that ships.

Implement a Jira issue, open a pull request, and report the work on the issue. Insert into the prompt of a Jira column rule that does the building.

Work on a Flutter/Dart repository (one with a pubspec.yaml) in the sandbox — resolve private pub dependencies, run analyze/test, and QA the UI on an Android emulator or by compiling to web, and screenshot it. Use whenever the repo you are working in is a Flutter app, and especially before reporting that you cannot see or exercise a mobile app's UI.

Route a file to the right reader by extension (pptx, docx, xlsx, pdf, and plain text). Use as the entry point when asked to read or summarize an uploaded file of unknown type.

Read and extract text, tables, and structure from .docx Word documents. Use when the user uploads or references a Word file to read or summarize.

Operate the user's deco Studio (decocms) organization from the terminal with their own login. Use when the user asks to read or change something in Studio, such as chats/threads, agents, the task board, connections, monitoring, or org files, or to send a message to a Studio agent.

Run Studio locally with chats and task runs executing in real Freestyle sandboxes, reachable through an ngrok tunnel. Use when asked to "run locally with Freestyle", "test claude-code chats locally", "run the sandbox harness end to end", or to try daemon/harness-runner changes in a real sandbox.

Searches, saves, and maintains a local document index through a local RAG MCP server. Use when user says "search my docs", "save this page", "read around that chunk", "sync my index", or invokes `npx mcp-local-rag`.

Decision rules for when to use MCP Qdrant-Indexer semantic search vs grep/literal file tools. Use this skill when starting exploration, debugging, or answering "where/why" questions about code.

Performs hybrid semantic/lexical search with neural reranking for codebase retrieval. Use for finding implementations, Q&A grounded in source code, and cross-session persistent memory.

Verify the flutter-mcp-toolkit install, run doctor preflight, troubleshoot connection issues. Use when the toolkit isn't responding or first-time setup.

Maintain mcp_flutter releases, CHANGELOG, version pins, docs, and CI. Use when cutting a release, editing CHANGELOG.md, bumping VERSION, running release-please, sync-skills, check-contracts, or updating install/docs for npx skills and flutter-mcp-toolkit init.

Maintains showcase/flutter_test_app and intentcall web targets (Chrome, web codegen, WebMCP bootstrap, web-showcase, webmcp verify, Chrome DevTools MCP). Use when editing web/index.html, agent_manifest.json, intentcall_webmcp.generated.js, web platform sync, Chrome dogfood, WebMCP modelContext, or agent WebMCP list/execute.

Maintains showcase/flutter_test_app macOS showcase, native intentcall hooks (codegen, app_links invoke), and VM MCP validation. Use when editing macOS Runner, intentcall_codegen.sh, macOS dogfood, Screen Recording capture, or comparing macOS parity to web WebMCP.

Migrate Flutter app code from removed MCPCallEntry to AgentCallEntry after the hard cut. Use when upgrading mcp_toolkit, fixing compile errors after a major bump, or running flutter-mcp-toolkit migrate agent-entries.

Read state from a running Flutter app — semantic snapshot, view details, errors, screenshots, VM info. Use when you need to understand what the app is showing.

Entry point for inspecting or driving a running Flutter app from your AI assistant — routes to the right task skill (inspect / control / debug / custom app surfaces) and runs preflight.

Runs and records showcase/flutter_test_app dogfood iterations (tool_quality_rubric, run_dogfood_eval.sh, dogfood_web_eval.yaml). Use when scoring MCP/intentcall quality, appending iteration N, comparing regressions, or CI static/weekly eval gates.

Diagnose problems in a running Flutter app — read logs, evaluate Dart expressions, interpret error envelopes. Use when something broke.

Use this skill when the agent exposes app-specific surfaces by registering custom MCP tools and resources inside the Flutter app (mcp_toolkit dynamic registry — AgentCallEntry, addEntries). Covers tool vs resource vs evaluate-expression, Map-based handlers, schema strictness, discovery via fmt_list_client_tools_and_resources, fmt_client_tool, fmt_client_resource, and lifecycle pitfalls.

Drive a running Flutter app — tap, scroll, type, fill forms, hot-reload, navigate. Use when you need to interact with the UI.

Use this skill when writing or running repeatable E2E scenarios for Flutter apps as checked-in Dart with the `flutter_mcp_harness` package (`packages/harness`) — build/launch the app, attach to its VM service, drive widgets (snapshot/tap/enter text) and assert without MCP; also covers the repo showcase launcher (`make showcase`, `--web`, `--stop`) and multi-instance desktop compositions. For interactive assistant-driven debugging through MCP `fmt_*` tools, use `flutter-mcp` instead.

Run Flutter MCP runtime validation from CLI in two steps (launch app, then run validate-runtime), including toolkit-extension gating, screenshot/layout capture, app error collection, optional reload verification, and retry handling for transient first-connect failures.

Generic contract/schema boundary audit across authoring, discovery, validation, and execute—detecting split-brain between listings and invoke paths, gateway divergence, and permissive placeholders. Use when changing tool registration, RPC/plugin registries, dynamic tools, MCP or WebMCP surfaces, CLI exec aliases, OpenAPI or JSON Schema contracts, migrators/codegen, bridge argument encoding, or platform docs that describe API contracts.

Use this skill when wiring automation for a Flutter app — choosing the right surface (interactive MCP `fmt_*` tools, CLI one-shots, programmatic `flutter_mcp_harness` Dart scenarios, the browser CDP tier, or an external build+lifecycle runner like oka), composing the chain end-to-end, or integrating `oka` (`oka init`/`dev`/`run`, device targets, the runner-session contract, `driverForLiveSession`). Also use when the user asks which tier or mode to use, how to drive/build their app from Dart or CI without a chat round-trip, or how oka and flutter-mcp-toolkit fit together. For authoring repeatable scenarios inside this repo's showcase, use `flutter-mcp-e2e-harness`; for interactive debugging, use `flutter-mcp`.

Use this skill whenever inspecting, interacting with, or live-editing a running Flutter app via the Flutter MCP toolkit server (`mcpServers` key **`flutter-mcp-toolkit`**, or legacy **`flutter-inspector`**). Covers preflight, snapshot/tap/enter/scroll loop, hot-reload validation, and error envelope parsing.

Turn the current conversation into a PRD and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices.

Generate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions "design it twice".

Turn a loose idea into a sequenced map of investigation tickets, then drive them to resolution one at a time.

Prepare or run a polished BlockRun trading demo that discovers a current Polymarket market, combines live price, probability history, smart-money, and liquidity evidence into a balanced signal, produces a real order dry-run, and verifies orders or positions. Use for live demos, signal-to-trade workflows, current crypto prediction markets, or when an agent must decide whether a candidate is safe and presentable before trading.

Use when the user wants to actually PLACE, manage, or redeem bets on Polymarket (not just read odds — that's blockrun_markets). Covers setup (deposit wallet, funding, approvals), buy/sell with confirm gating, positions, redeeming winnings, geoblock handling, and the end-to-end demo flow.

Automate and test Flutter applications — launch apps, inspect widgets, tap elements, enter text, scroll, swipe, take screenshots, validate state, and debug via Dart VM Service Protocol. Use when the user wants to run Flutter app tests, automate Flutter UI interactions, inspect widget trees, debug a running Flutter app, or perform gesture-based testing.

Open the Cross-Code Organizer (CCO) dashboard — view and manage all memories, skills, MCP servers, hooks, and configs across scopes

Inspect tfmcp Rust duplicate-code candidates with similarity-rs and judge whether sharing their responsibilities would improve maintenance.

Verify tfmcp MCP lifecycle, transports, and local Terraform operations end to end with real implementations and isolated fixtures.

Run and interpret cargo-coupling analysis for tfmcp, including coupling health, hotspots, and baseline comparisons.

AI Team OS 里用 CC 内置 Workflow 的两件事：产出回写 OS 的标准模板（§1-2、§4），以及审查分级与派工档位纪律（§3/§3.1）。准备调用 Workflow 编排子 agent 时，或要判定一件事该按 L0/L1/L2 哪一档审查（含「这活要不要开 workflow」）时使用。

发布 AI Team OS 新版本的完整清单——预检、版本七处锁步、中英双语 CHANGELOG、双份 dist 构建、私有术语扫描、commit/tag、双仓推送、建 GitHub Release 条目并核对 latest 徽章、事后核对。当准备发版、补建漏掉的 Release 条目、或核对已发版本的线上状态时使用。

CC 与 Codex（或两个各自独立运行的 CC 会话）之间用 OS 信道留言、查未读、被新消息唤醒。触发：要跨 harness 把结论交给对端、给对端留言或清未读、用户问"能不能让 Claude 和 Codex 互相沟通"。不适用于派子 agent 或 workflow 内部协作——那是 Agent 与 SendMessage，信道对同会话的子 agent 没有意义。

你被派来参加 AI Team OS 会议（派你的 prompt 里带 meeting_id），且要在 Round 2 及之后发言时用：怎么拿到自己的 agent_id、怎么引用前人发言、round_number 怎么填。Round 1 的材料、发言规则与调用样例已经写在派你的 prompt 里，不必读本技能。

用 meeting_* 工具跑一场多 Agent 会议：创建 → 亲自 spawn 参与者 → 推进轮次 → 签到 → conclude 并把决策上墙。仅当已决定以「开会」形式产出一个需上墙的决策时使用；典型场景是减法类提案：回退既有设计决策、砍工具删表、削弱或删除自动检查——这类结论必须留痕。加法（新增检查、加严判据、新增工具）不需要开会，写清理由直接做。用户只是让你评审代码、复盘、比较方案时不要触发——那是直接做，或派一个审查 agent。

Inspect, edit, verify, save, and export an open Adobe Premiere Pro project through the premiere-pro MCP server. Use for rough cuts, timeline assembly or cleanup, clip and track changes, transitions and effects, dialogue or audio adjustments, captions, clip metadata and XMP, project organization, frame inspection, and delivery exports.