원클릭으로
pipeline
End-to-end source processing: reduce → reflect → reweave → verify. Fresh context per phase for optimal LLM attention.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
End-to-end source processing: reduce → reflect → reweave → verify. Fresh context per phase for optimal LLM attention.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Add a new knowledge domain to an existing vault. Creates domain-specific MOCs, templates, and vocabulary mappings.
Research-backed vault evolution guidance. Toggle features (self-space, semantic search), rebalance MOCs, adjust dimensions with full rationale.
Query the Ars Contexta research methodology. Answers 'why does my system do X?' with research-backed explanations from 249 interconnected claims.
Analyze the vault's knowledge graph. Subcommands: orphans, density, bridges, synthesis, traverse, clusters, stats.
Run vault health diagnostics. 3 modes: quick (schema+orphans+links), full (all 8 categories), three-space (boundary violations).
Contextual guidance and command discovery. Shows available skills, agents, and intelligent suggestions based on vault state.
| name | pipeline |
| description | End-to-end source processing: reduce → reflect → reweave → verify. Fresh context per phase for optimal LLM attention. |
| tags | ["processing","automation","pipeline"] |
Delegates to: @ars-contexta:processor
Invoke this skill to run the complete source-to-knowledge pipeline in a single command. Composes reduce, reflect, reweave, and verify in sequence.
<file> — run the full pipeline for a specific source fileall — run the full pipeline for all files currently in inbox/If no argument is provided, list inbox/ contents and ask which sources to process.
Execute the four phases in strict sequence. Each phase runs as a separate task delegation to ensure fresh LLM context — this prevents attention degradation over long processing runs and keeps each phase focused on its specific concern.
Phase 1: Reduce Delegate to reduce skill with the specified source argument.
Extract insights from source material. Create atomic notes with prose titles, YAML frontmatter, source attribution, and initial topic suggestions. Move processed sources to archive/.
Wait for completion. Collect the list of created note paths before proceeding.
Phase 2: Reflect
Delegate to reflect skill with recent (the notes just created in Phase 1).
Find connections between the new notes and the existing vault. Add typed Relevant Notes sections. Update MOCs. Flag synthesis opportunities.
Wait for completion. Collect the list of backward-link targets before proceeding.
Phase 3: Reweave
Delegate to reweave skill with recent (the connections made in Phase 2).
Update older notes with backward links to the new notes. Update descriptions where new context changes the summary. Flag MOC rebalancing needs.
Wait for completion before proceeding.
Phase 4: Verify
Delegate to verify skill with recent (all notes touched in Phases 1-3).
Run schema compliance, description quality, link health, and topics coverage checks on all notes created or modified during this pipeline run.
Phase boundary reporting After each phase, output a brief status line before starting the next:
Phase 1 (reduce): {n} notes created from {n} sources — done
Phase 2 (reflect): {n} connections added, {n} MOCs updated — done
Phase 3 (reweave): {n} backward links added, {n} descriptions updated — done
Phase 4 (verify): {n} PASS, {n} WARN, {n} FAIL — done
Final summary
Pipeline complete — {YYYY-MM-DD}
Sources processed: {n}
Notes created: {n}
Connections made: {n}
Issues found: {n} (see verify output above)
If any verify phase produces FAIL results, surface the specific issues and suggest next steps. Do not block on warnings.