원클릭으로 Manus에서 모든 스킬 실행

atom-concepts-backfill

스타0

포크1

업데이트2026년 6월 12일 18:09

Bridge an upstream gap: extract_atoms never stamps `frontmatter.concepts` on page-derived atoms, but synthesize_concepts consumes ONLY that field — so concept synthesis is permanently skipped ("no atoms with concept refs"). This script batch-labels atoms lacking the field with 1-3 kebab-case shared-vocabulary topic labels (sonnet) and stamps them, so upstream synthesize_concepts can tier + materialize concepts/<label> pages. Idempotent: only touches atoms WITHOUT `concepts`; safe to rerun after each extract_atoms drain. Retires when the upstream fix lands.

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

ChenyqThu

ChenyqThu/jarvis-knowledge-os-v2

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

atom-concepts-backfill — bridge for the extract_atoms → synthesize_concepts gap

上游断点（2026-06-11 定位，gbrain v0.42.37.0/v0.42.38.0）

src/core/cycle/synthesize-concepts.ts 头注释（设计意图）："when the Haiku 3-check from extract_atoms decides 'this atom is about concept X', it stamps the field" — :92 只消费 frontmatter.concepts。
src/core/cycle/extract-atoms.ts 的 page 提取路径从不写该字段（只写 type/atom_type/lesson/source_hash/source_quote/virality_score/ emotional_register）。
结果：所有 page-derived atoms 永远进不了 concept 合成（生产实证：696 atoms / 0 含 concepts / synthesize_concepts 连续 skip）。
上游 issue 已起草（见 TODO）；本脚本为 fork 侧桥接，上游修复后自动退场（届时新 atoms 自带 concepts，本脚本查询恒为空集）。

用法

# 必须从 repo root 跑（bun 自动加载 .env.local 的 crs 代理 key/base）。
# Claude Code shell 注入的 ANTHROPIC_BASE_URL 会覆盖 .env.local 并破坏
# key/base 配对 —— 先 unset（见 memory: pitfall-anthropic-base-url）。
unset ANTHROPIC_BASE_URL

bun run skills/kos-jarvis/atom-concepts-backfill/run.ts              # dry-run（只打标签分布，不写库）
bun run skills/kos-jarvis/atom-concepts-backfill/run.ts --limit 60 --apply   # 试点
bun run skills/kos-jarvis/atom-concepts-backfill/run.ts --apply     # 全量

默认 dry-run；--apply 才写库（jsonb merge concepts + concepts_by 审计标记，不动 body/chunks/embedding，无 re-embed 开销）。
--limit N / --batch N（默认 40/批）；KOS_CONCEPTS_MAX_SPEND 花费上限（默认 $5）；KOS_CONCEPTS_MODEL 模型覆盖（默认 claude-sonnet-4-6，按 Lucien 2026-06-09 路由：上游设计的 Haiku 档 → sonnet）。
词表跨批次累积并从已 stamp 的 atoms 预热，保证标签聚类（ synthesize_concepts 需要 ≥2 atoms 同标签才物化 T3 concept 页）。

验证回路

--limit 60 --apply 试点 → 看输出的 "cluster(s) with >=2 atoms"。
unset ANTHROPIC_BASE_URL && ~/.bun/bin/gbrain dream --phase synthesize_concepts --dir ~/brain → 确认 concepts/<label> 页出现（SELECT slug FROM pages WHERE frontmatter->>'synthesized_by' LIKE 'synthesize_concepts%'）。
全量 --apply；每次 extract_atoms drain 推进后重跑一次补新增。

이 저장소의 다른 Skills

같은 저장소

orphan-reducer

ChenyqThu/jarvis-knowledge-os-v2

Classify vector-similar pages against each orphan page via Haiku 4.5 and add typed inbound references (link_type="related", relation encoded in context). Goal: chip away at the 1803/1952 orphan pile (v1 wiki flat import) across multiple sweep runs. Default is --dry-run; --apply does DB link inserts + markdown sentinel upsert on candidate pages that exist on disk, then commits ~/brain changes.

2026-07-140

entity-dedup

ChenyqThu/jarvis-knowledge-os-v2

Merge WITHIN-SOURCE duplicate entity-variant pages (e.g. people/lucien + people/lucien-chen, companies/tp-link-system-inc + tp-link-systems-inc) into one canonical node: repoint links + facts + display-name aliases, register a slug redirect, soft-delete the alias pages. An LLM disambiguates each cluster (merge / ambiguous / distinct); ambiguous first-name-vs-multiple-full-names cases are quarantined, never merged. Default --dry (simulate + report, zero writes). Cross-source pages are NEVER touched — the two-axis design keeps them legitimately separate.

2026-07-130

enrich-sweep

ChenyqThu/jarvis-knowledge-os-v2

One-shot sweep that reads every page in the brain, extracts every person/company/concept/project mentioned via Haiku NER, dedupes, tiers by mention count + context, and creates stub pages for entities that lack one. Tier 2 candidates get Tavily web-search augmentation; Tier 3 stays brain-only (confidence: low). This is the primary G1 payoff skill from JARVIS-NEXT-STEPS.md §5.

2026-06-260

kos-patrol

ChenyqThu/jarvis-knowledge-os-v2

Daily sweep: runs kos-lint + staleness detection + knowledge-gap analysis, then produces a MEMORY-format digest ready to be appended to OpenClaw's MEMORY.md by the digest-to-memory skill (Sun 22:00 weekly). Also writes a dashboard to ~/brain/agent/dashboards/knowledge-health-<date>.md.

2026-06-150

corpus-ingest

ChenyqThu/jarvis-knowledge-os-v2

One-command bulk-ingest pipeline (D→I→K→W) for a single isolated source. Sequences the 7 fork stages — register → ingest → embed → enrich → graph → (per-entity synth, opt-in) → corpus-synth — over a folder of frontmatter'd .md files, each idempotent/incremental so re-runs only do new work. Turns a systematic corpus (product docs, FAQ, a research library) into a queryable knowledge asset: facts (RAG) + entities + graph + corpus-level viewpoints.

2026-06-020

corpus-synth

ChenyqThu/jarvis-knowledge-os-v2

Corpus-level Wisdom synthesis. Where synthesis-sweep deepens ONE entity, corpus-synth synthesizes a whole imported knowledge base into queryable "观点与洞察" pages — the corpus's major themes, cross-document tensions, transferable patterns, and open questions. Top-down: Opus reads a corpus map, proposes the themes, then deeply synthesizes each. Turns a freshly imported systematic KB into knowledge AND wisdom (not just RAG facts).

2026-06-020

name	atom-concepts-backfill
version	1.0.0
description	Bridge an upstream gap: extract_atoms never stamps `frontmatter.concepts` on page-derived atoms, but synthesize_concepts consumes ONLY that field — so concept synthesis is permanently skipped ("no atoms with concept refs"). This script batch-labels atoms lacking the field with 1-3 kebab-case shared-vocabulary topic labels (sonnet) and stamps them, so upstream synthesize_concepts can tier + materialize concepts/<label> pages. Idempotent: only touches atoms WITHOUT `concepts`; safe to rerun after each extract_atoms drain. Retires when the upstream fix lands.
triggers	["atom concepts backfill","backfill atom concepts","concepts backfill"]
mutating	true

atom-concepts-backfill — bridge for the extract_atoms → synthesize_concepts gap

上游断点（2026-06-11 定位，gbrain v0.42.37.0/v0.42.38.0）

src/core/cycle/synthesize-concepts.ts 头注释（设计意图）："when the Haiku 3-check from extract_atoms decides 'this atom is about concept X', it stamps the field" — :92 只消费 frontmatter.concepts。
src/core/cycle/extract-atoms.ts 的 page 提取路径从不写该字段（只写 type/atom_type/lesson/source_hash/source_quote/virality_score/ emotional_register）。
结果：所有 page-derived atoms 永远进不了 concept 合成（生产实证：696 atoms / 0 含 concepts / synthesize_concepts 连续 skip）。
上游 issue 已起草（见 TODO）；本脚本为 fork 侧桥接，上游修复后自动退场（届时新 atoms 自带 concepts，本脚本查询恒为空集）。

用法

# 必须从 repo root 跑（bun 自动加载 .env.local 的 crs 代理 key/base）。
# Claude Code shell 注入的 ANTHROPIC_BASE_URL 会覆盖 .env.local 并破坏
# key/base 配对 —— 先 unset（见 memory: pitfall-anthropic-base-url）。
unset ANTHROPIC_BASE_URL

bun run skills/kos-jarvis/atom-concepts-backfill/run.ts              # dry-run（只打标签分布，不写库）
bun run skills/kos-jarvis/atom-concepts-backfill/run.ts --limit 60 --apply   # 试点
bun run skills/kos-jarvis/atom-concepts-backfill/run.ts --apply     # 全量

默认 dry-run；--apply 才写库（jsonb merge concepts + concepts_by 审计标记，不动 body/chunks/embedding，无 re-embed 开销）。
--limit N / --batch N（默认 40/批）；KOS_CONCEPTS_MAX_SPEND 花费上限（默认 $5）；KOS_CONCEPTS_MODEL 模型覆盖（默认 claude-sonnet-4-6，按 Lucien 2026-06-09 路由：上游设计的 Haiku 档 → sonnet）。
词表跨批次累积并从已 stamp 的 atoms 预热，保证标签聚类（ synthesize_concepts 需要 ≥2 atoms 同标签才物化 T3 concept 页）。

验证回路

--limit 60 --apply 试点 → 看输出的 "cluster(s) with >=2 atoms"。
unset ANTHROPIC_BASE_URL && ~/.bun/bin/gbrain dream --phase synthesize_concepts --dir ~/brain → 确认 concepts/<label> 页出现（SELECT slug FROM pages WHERE frontmatter->>'synthesized_by' LIKE 'synthesize_concepts%'）。
全量 --apply；每次 extract_atoms drain 推进后重跑一次补新增。