菜单
CRITICAL: MUST run for EVERY message. Detects agent, complexity, AND model automatically. Without this, tasks route to wrong agents and use wrong models, degrading quality and wasting tokens.
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。
基于 SOC 职业分类
| name | agent-detector |
| description | CRITICAL: MUST run for EVERY message. Detects agent, complexity, AND model automatically. Without this, tasks route to wrong agents and use wrong models, degrading quality and wasting tokens. |
| autoInvoke | true |
| priority | highest |
| model | haiku |
| triggers | ["every message","always first"] |
| allowed-tools | NONE |
| user-invocable | false |
AI-consumed reference. Optimized for Claude to read during execution. Human-readable explanation: see docs/architecture/HIERARCHICAL_PLANNING.md or docs/getting-started/ depending on topic.
Runs FIRST for every message.
complexity[4]{level,criteria,approach}:
Quick,"Single file / simple fix / clear scope","Direct implementation"
Standard,"2-5 files / feature / some unknowns","Scout then implement"
Deep,"6+ files / architecture / vague scope","run-orchestrator"
Project,"Multi-feature / multi-session / weight ≥ 3 on bridge heuristic AND no active plan","/aura-frog:plan bootstrap then per-task /run anchored"
Project (v3.7.2+): Emitted when rules/workflow/run-plan-bridge.md triggers sum to weight ≥ 3 AND .claude/plans/active.json is absent. run-orchestrator Step 0 owns the user prompt (plan / deep / details) and the scratch-file handoff. Otherwise Quick/Standard/Deep classification is unchanged.
Prefer the session model — inherit it. The model the user launched with is the signal of their intent and budget; use it for all substantive work (Standard / Deep / architecture / planning / build / review). Do NOT force-upgrade to a named model (e.g. Opus) for "complex" — if the user wanted more capability they'd run it, and naming a model means a newer/stronger one is ignored.
The only deliberate override is down-shifting to haiku for trivial mechanical work (classification, detection, state bookkeeping) where a wrong answer costs little. So: Quick/classification → haiku; everything else → inherit (session model). See rules/core/small-to-large-routing.md.
Primary ≥80 (leads), Secondary 50-79 (supports), Optional 30-49, Skip <30. tester: Always secondary unless explicit test request.
.claude/cache/agent-detection-cache.json — reuse within workflow (phase >1). Invalidate on new workflow, phase 1, or user override.
5 techniques for different problem types. Use when stuck or facing complex challenges.
Structured thinking process for complex analysis. Supports revision, branching, and dynamic adjustment.
Angular 17+ gotchas and decision criteria. Covers signals vs observables, standalone patterns, and common pitfalls Claude gets wrong.
Designs RESTful APIs with endpoint naming, versioning strategies (URL path, header-based), pagination (offset and cursor), error response schemas, and OpenAPI conventions. Use when the user asks about REST API design, creating endpoints, URL structure, API versioning, status codes, Swagger, or OpenAPI specs.
Fast bug fixes with root cause investigation + TDD. Enforces 'no fix without root cause' discipline and verification protocol. Without this skill, fixes are applied at symptoms instead of sources, and bugs return.
Draft → generate verification questions → answer independently via tools → revise. Catches hallucinated facts in reports and reviews. MANDATORY for Phase 4 security/test claims. Paper: Dhuliawala et al. 2023.