菜单
Exhaustive search mode. Use when you need broad codebase or documentation coverage, multiple parallel searches, and a synthesized inventory of findings.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Interview the user relentlessly about a plan or design until branch-level decisions are resolved for execution.
Access Figma designs, extract design systems, and retrieve component specifications. Use when implementing UI from Figma mockups, extracting design tokens, or analyzing design files.
Enforce cost-aware MCP usage. Use when a task might trigger heavy external tools, web search, or broad context expansion. Prevents token burn by ensuring MCPs are only used when local context is insufficient.
Navigate the Warmplane mcp0 facade efficiently. Use when the active config exposes provider capabilities through `mcp0_*` tools and you need to discover or call provider tools without brute-force describing large capability sets. Trigger on requests involving mcp0, Warmplane, or provider work through the facade such as Linear, Notion, Figma, New Relic, Context7, grep.app, or Storybook tools.
Use this when the user needs to control Chrome, navigate to a page, inspect a tab, click or fill elements, take screenshots, or automate a browser flow with aeroxy/chrome-devtools-cli.
Guidelines for creating and managing implementation plans with citations
基于 SOC 职业分类
| name | search-mode |
| description | Exhaustive search mode. Use when you need broad codebase or documentation coverage, multiple parallel searches, and a synthesized inventory of findings. |
ACTIVATION: When loaded, immediately maximize search effort across all available tools.
MAXIMIZE SEARCH EFFORT
Launch multiple background agents in parallel only when the extra coverage is worth the context and wall-clock cost.
// Codebase patterns (INTERNAL)
task(agent="explore", prompt="Find [pattern 1]...", background=true)
task(agent="explore", prompt="Find [pattern 2]...", background=true)
task(agent="explore", prompt="Find [pattern 3]...", background=true)
// External resources (EXTERNAL)
task(agent="researcher", prompt="Find docs for...", background=true)
task(agent="researcher", prompt="Find GitHub examples...", background=true)
Use the explore agent's scope-first tool hierarchy for local search, then add external research only when the task crosses into documentation or OSS examples.
For: "Find all authentication implementations"
// Parallel agents
task(agent="explore", prompt="Find auth middleware implementations", background=true)
task(agent="explore", prompt="Find login/logout functions", background=true)
task(agent="explore", prompt="Find JWT/session handling", background=true)
task(agent="researcher", prompt="Find NextAuth.js best practices", background=true)
// Direct tools (in parallel)
glob(pattern="**/*auth*")
grep(pattern="authenticate|authorization|jwt|session", include="*.ts")
ast_grep_search(pattern="async function $NAME($$$) { $$$ }", lang="typescript")
lsp_symbols(query="authenticate", scope="workspace")
Synthesize ALL findings into: