formated-results
Closing skill for the research-executor, loaded as the last step of formated-specs. Summarize the design just produced into one research-result JSON fenced…
SOP: Turn the feeding plan into the compiler''s $ARGUMENTS and run the external ARA compiler once inline to produce ../ara/
$ npx -y skills add yogsoth-ai/de-anthropocentric-research-engine --skill ara-compile --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/ara-compileContext preview
The summary Claude sees to decide when to auto-load this skill.
SOP: Turn the feeding plan into the compiler''s $ARGUMENTS and run the external ARA compiler once inline to produce ../ara/
name: ara-compile description: 'SOP: Turn the feeding plan into the compiler''s $ARGUMENTS and run the external ARA compiler once inline to produce ../ara/' version: 1.0.0 category: ara-from-context type: sop campaign: ara-from-context input: Completed feeding plan + aligned 大方向 (from north-star-align) output: A populated ara/ directory (sibling to context/), ARA Seal Level 1 passed dependencies: skills: - compiler
**Key question**: 怎么把投喂计划喂给外部 compiler,一次抽出一份内部一致的 ARA?
先确认外部 `compiler` skill 可 load(ARA skills 已装:`npx @ara-commons/ara-skills`)。 若不可用,提示用户安装并**停下**,不要静默继续。
ARA 的 cross-layer binding(claim→proof→evidence、tree→claim)必须**全局一致**。 分批 compile 会各自从 C01 起撞 ID、断 tree,汇总等于重缝半成品 —— 正是 ARA 要消灭 的事。compiler 自带覆盖度循环(max 3 轮)+ 内建 `Task` 工具;真需要并行由它**内部** 自理,本 SOP 不越俎拆分。
1. **把投喂计划整理成 compiler 的 `$ARGUMENTS`**:
例:
compiler context/2026-06-06-01-30-stage7-...md context/2026-06-05-...stage6...md \
context/figures/*.png \
--output ../ara/ \
主干=stage7(报告线);其余为过程线/图片;大方向:<从 north-star-align 来的一段>2. **一次 inline 运行**:`Skill` load **compiler**,传上面的 `$ARGUMENTS`。 compiler 跑 4 阶段(语义解构 → 认知映射 → src 层 → 探索图抽取)+ 覆盖度循环
3. **Seal Level 1 不过**:compiler 自带 fix-iterate(2–3 轮),本 SOP 不接管; 若仍不过,把失败报告**透传**给用户,停。
`<workspace>/ara/`(`logic/ src/ trace/ evidence/ PAPER.md`),Level 1 已过。 交给 `ara-rigor-review`。
The complete research orchestration system for AI-native science. What It Does Design Philosophy Architecture (v3.2.2) Quick Start Configuration Roadmap License DARE is not a tool that helps you do research. It is the researcher.
Repo: yogsoth-ai/de-anthropocentric-research-engine
Closing skill for the research-executor, loaded as the last step of formated-specs. Summarize the design just produced into one research-result JSON fenced…
Spec-slot skill for the research-executor. Emit the 4-layer DARE orchestration of the assigned topic as one research-graph JSON fenced block in your reply.…
Loss-1 judge (codex role). Given one sample's de-identified dialogue and its PolicyCard, decide axis-by-axis whether the user-simulator enacted the card's…
Loss-2 judge (codex role). Over one topic's 6 shuffled research-design samples, pairwise-rank by quality using the D1–D5 standard. Emit the pairwise log; the…
The optimizer brain for the ladder-foundry pretraining loop. Runs the two-level nested batch loop, delegates gating to gate_eval, attributes a failing batch to…
Tactic: Extract atomic units from one paper and score how much of a caller-supplied summary covers. Use for ACU-style binary or Nugget-style ternary recall…