菜单
Build an RKC ContextPack. Structural max_hops=2 plus Finding→Claim→Evidence spine. Question roots allow one inbound hop on answers.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/SpillwaveSolutions/research-knowledge-capture --skill research-pack命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Archive AI deep-research dumps and articles into RKC source-assets. Hash, copy, write SourceDocument + ResearchTask shells. Optionally run the Phase 2 extractor. Idempotent on bytes plus prompt hash plus extractor version.
Create missing Subject and ResearchArea nodes and write has_task / has_subject on an existing RKC tree. Idempotent repair for 0.2.0 archive-only ingest.
Bind a Cursor agent to RKC. Capture research into OKF. Do not project the graph.
正在显示 SKILL.md
| name | research-pack |
| description | Build an RKC ContextPack. Structural max_hops=2 plus Finding→Claim→Evidence spine. Question roots allow one inbound hop on answers. |
python3 ${CLAUDE_PLUGIN_ROOT}/scripts/rkc_pack.py <id> --root knowledge
max_hops default 2.asserts → Claim and evidenced_by → Evidence do not consume hops.answers.max_nodes default 20. Truncate by rank: status → verified → confidence → newer as_of → id.Do not cite Chroma or Kuzu blobs. Citations are Finding → Claim → Evidence → source-asset.