菜单
Generates or refreshes research artifacts and RLM evidence for a ticket. Use when research stage should produce or update canonical RLM outputs. Do not use when the request is initial ticket bootstrap in `idea-new` or implementation planning in `plan-new`.
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
Aggregates shared runtime navigation and cross-skill ownership links. Use when resolving shared runtime entrypoints or ownership boundaries. Do not use when the request belongs to `aidd-policy` output/question rules or `status` summary reporting.
Owns shared DocIO runtime for markdown slicing/patching, actions validation/apply, and context-map expansion. Use when stage skills need canonical DocIO operations. Do not use when the request primarily belongs to `aidd-flow-state` lifecycle updates or `aidd-core` ownership routing.
Owns shared flow/state runtime for stage state, progress/tasklist, and stage-result lifecycle. Use when orchestration needs canonical active-stage, progress, tasklist, or stage-result commands. Do not use when the request is about `aidd-docio` action patch mechanics or `status` read-only summaries.
Initializes AIDD workspace scaffolding for the current project. Use when bootstrapping canonical `aidd/` templates and project-owned test execution contract in `aidd/config/gates.json`. Do not use when the request is to start feature ideation (`idea-new`) or to inspect existing ticket progress (`status`).
Enforces loop-mode discipline for implement/review/qa (packs, scope, no questions). Use when loop stages require bounded scope and retry policy. Do not use when direct stage execution belongs to `implement`, `review`, or `qa` stage skills.
Owns shared observability/runtime reporting for diagnostics, inventory, identifiers, logs, and DAG export. Use when deterministic diagnostics or inventory exports are required. Do not use when request scope is `status` stage summaries or `aidd-core` runtime ownership mapping.
| name | researcher |
| description | Generates or refreshes research artifacts and RLM evidence for a ticket. Use when research stage should produce or update canonical RLM outputs. Do not use when the request is initial ticket bootstrap in `idea-new` or implementation planning in `plan-new`. |
| argument-hint | $1 [note...] [--paths path1,path2] [--keywords kw1,kw2] [--note text] |
| lang | en |
| prompt_version | 1.2.35 |
| source_version | 1.2.35 |
| allowed-tools | ["Read","Edit","Write","Glob","Bash(rg *)","Bash(sed *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/researcher/runtime/research.py *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/set_active_feature.py *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/set_active_stage.py *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/tasks_derive.py *)"] |
| model | inherit |
| disable-model-invocation | true |
| user-invocable | true |
Follow feature-dev-aidd:aidd-core.
research.python3 ${CLAUDE_PLUGIN_ROOT}/skills/researcher/runtime/research.py --ticket <ticket> --auto.--paths, --keywords, --note) when targeted refresh is needed.*-rlm-targets.json, *-rlm-manifest.json, *-rlm.worklist.pack.json, optional *-rlm.pack.json).feature-dev-aidd:researcher. First action: read RLM pack/worklist.--auto mode run bounded canonical finalize recovery once (rlm_finalize --bootstrap-if-missing) from stage runtime.python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-rlm/runtime/rlm_finalize.py --ticket <ticket>) and append research handoff via python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/tasks_derive.py --source research --append./feature-dev-aidd:plan-new <ticket> only when RLM readiness is ready and the research document is synchronized with the current evidence.ready, return the deterministic blocker or pending output instead of guessing a downstream handoff.python3 ${CLAUDE_PLUGIN_ROOT}/skills/.../runtime/...)./skills/... runtime paths are forbidden; use only ${CLAUDE_PLUGIN_ROOT}/skills/.../runtime/....python3 ${CLAUDE_PLUGIN_ROOT}/skills/researcher/runtime/research.py--ticket <ticket> with optional path/keyword/note overrides.rlm_status, rlm_pack_status).reason_code + next_action.AIDD:ACTIONS_LOG: n/a.python3 ${CLAUDE_PLUGIN_ROOT}/skills/researcher/runtime/research.py --ticket <ticket> --auto.rlm_status remains pending after bounded auto recovery, hand off to shared owner:
python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-rlm/runtime/rlm_finalize.py --ticket <ticket>.aidd/docs/research/<ticket>.md; why: keep artifact shape aligned with canonical workspace template).rlm_status is not ready or RLM pack is missing; why: run canonical shared RLM API outside stage-local orchestration).