[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C21RNDW6P4ENRJBHGD95CX":3,"skill-same:01M6C21RNDW6P4ENRJBHGD95CX":83},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":26,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":59,"commitSha":60,"skillDirectory":61,"forksCount":62,"repositoryHtmlUrl":63,"repositoryHomepageUrl":64,"ownerHtmlUrl":65,"ownerBlogUrl":66,"ownerDescription":67,"updatedAt":68,"install":69,"resources":82},"01M6C21RNDW6P4ENRJBHGD95CX","research-workflow",null,"Execute multi-stage research workflows","jmagly\u002Faiwg","jmagly","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1159087?v=4","01M6C10AK8N2WCSQN11FYGTJWN","Joseph Magly","01M6C10AQ8A2RMED1Y1BSVY7NF",229,"MIT","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fframeworks\u002Fresearch-complete\u002Fskills\u002Fresearch-workflow\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"workflow","research","research workflow","papers","with","multi","input","name","stage","Citation","me","build","execute","plan","start","help","workflows","2026-11-10 10:12:30",false,"---\nnamespace: aiwg\nname: research-workflow\nplatforms: [all]\ndescription: Execute multi-stage research workflows\ntriggers:\n  - \"help with research papers\"\n  - \"help me build a cited research corpus\"\n  - \"start a research workflow or citation plan\"\n  - \"research workflow\"\ncommandHint:\n  argumentHint: \"[workflow-name] [--input parameters] [--stage n]\"\n  category: research-workflows\n---\n\n# Research Workflow Command\n\nEvery acquisition, induction, and synthesis dispatch uses the shared [context engineering contract](..\u002F..\u002F..\u002F..\u002F..\u002F..\u002Fdocs\u002Fcontext-engineering-contract.md) and [verification contract](..\u002F..\u002F..\u002F..\u002F..\u002F..\u002Fdocs\u002Fverification-contracts.md). Synthesis completion requires citation resolution, GRADE-calibrated claims, and evidence\u002Finference\u002Frecommendation separation.\n\nExecute complete multi-stage research workflows from discovery through archival.\n\n## Instructions\n\nWhen invoked, orchestrate multi-agent research workflows:\n\n1. **Load Workflow Definition**\n   - Identify workflow by name or load custom workflow YAML\n   - Parse stages, agents, dependencies\n   - Validate workflow structure\n\n2. **Execute Stages Sequentially**\n   - Each stage invokes specific agents\n   - Pass outputs from one stage to next\n   - Handle stage failures and retries\n   - Track progress and status\n\n3. **Monitor Execution**\n   - Display progress indicators\n   - Log all agent invocations\n   - Capture intermediate outputs\n   - Track resource usage (tokens, time)\n\n4. **Handle Gates**\n   - Pause for human approval at designated gates\n   - Present artifacts for review\n   - Collect feedback and decisions\n   - Resume or abort based on input\n\n5. **Generate Report**\n   - Summarize workflow execution\n   - Report outcomes for each stage\n   - Calculate quality metrics\n   - Archive workflow state\n\n## Built-in Workflows\n\n| Workflow | Stages | Description |\n|----------|--------|-------------|\n| `discovery-to-corpus` | 5 | Full pipeline from search to documented findings |\n| `paper-acquisition` | 3 | Download, extract metadata, create finding document |\n| `quality-assessment` | 4 | GRADE assessment with citation validation |\n| `corpus-maintenance` | 6 | Periodic corpus health checks and updates |\n| `synthesis-report` | 4 | Generate synthesis report from topic cluster |\n| `citation-audit` | 3 | Validate all citations across corpus |\n\n## Arguments\n\n- `[workflow-name]` - Workflow to execute (required)\n- `--input [yaml-file]` - Input parameters for workflow\n- `--stage [n]` - Start from specific stage (default: 1)\n- `--pause-at [stage]` - Pause after specific stage\n- `--interactive` - Prompt for confirmation at each stage\n- `--dry-run` - Preview workflow without execution\n- `--resume [workflow-id]` - Resume previously interrupted workflow\n\n## Workflow Definitions\n\n### discovery-to-corpus\n\nComplete pipeline from literature search to documented findings:\n\n**Stages:**\n\n1. **Discovery** (agent: discovery-agent)\n   - Search academic databases for query\n   - Rank and filter results\n   - Present top candidates\n\n2. **Acquisition** (agent: research-acquisition-agent)\n   - Download selected papers\n   - Extract metadata\n   - Generate frontmatter\n   - Create finding documents\n\n3. **Documentation** (agent: documentation-agent)\n   - Parse PDFs\n   - Extract key findings\n   - Assess AIWG relevance\n   - Generate literature notes\n\n4. **Quality Assessment** (agent: quality-agent)\n   - Apply GRADE framework\n   - Calculate quality level\n   - Generate assessment reports\n   - Update frontmatter\n\n5. **Archival** (agent: archival-agent)\n   - Create BagIt packages\n   - Update fixity manifest\n   - Register in archival index\n\n**Human Gates:**\n- After Discovery: Select papers to acquire\n- After Quality Assessment: Approve quality levels\n\n### paper-acquisition\n\nStreamlined acquisition workflow:\n\n**Stages:**\n\n1. **Download** (agent: research-acquisition-agent)\n   - Fetch PDF from source\n   - Verify file integrity\n\n2. **Metadata Extraction** (agent: research-acquisition-agent)\n   - Parse PDF metadata\n   - Enrich via CrossRef\u002FSemantic Scholar\n   - Assign REF-XXX identifier\n\n3. **Document Creation** (agent: documentation-agent)\n   - Generate finding document from template\n   - Populate frontmatter\n   - Add placeholder sections\n\n### quality-assessment\n\nComprehensive quality assessment workflow:\n\n**Stages:**\n\n1. **GRADE Assessment** (agent: quality-agent)\n   - Determine baseline quality\n   - Apply downgrade\u002Fupgrade factors\n   - Calculate final GRADE level\n\n2. **Hedging Analysis** (agent: quality-agent)\n   - Generate appropriate hedging language\n   - Document forbidden phrases\n   - Create citation templates\n\n3. **Citation Validation** (agent: citation-agent)\n   - Scan corpus for citations of this source\n   - Check hedging compliance\n   - Generate remediation suggestions\n\n4. **Report Generation** (agent: quality-agent)\n   - Create assessment report\n   - Update frontmatter\n   - Save assessment YAML\n\n## Examples\n\n```bash\n# Execute full discovery-to-corpus workflow\n\u002Fresearch-workflow discovery-to-corpus --input discovery-params.yaml\n\n# Acquire specific paper\n\u002Fresearch-workflow paper-acquisition --input '{\"doi\": \"10.48550\u002FarXiv.2308.08155\"}'\n\n# Run quality assessment\n\u002Fresearch-workflow quality-assessment --input '{\"ref_id\": \"REF-022\"}'\n\n# Interactive mode with pauses\n\u002Fresearch-workflow discovery-to-corpus --interactive\n\n# Dry run to preview\n\u002Fresearch-workflow corpus-maintenance --dry-run\n\n# Resume interrupted workflow\n\u002Fresearch-workflow resume wf-20260203-123456\n```\n\n## Expected Output\n\n```\nExecuting Workflow: discovery-to-corpus\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nInput Parameters:\n  Query: \"agentic workflows for software development\"\n  Max results: 10\n  Year from: 2020\n\nWorkflow Progress: [████░░░░░░] Stage 1\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 1: Discovery (agent: discovery-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n  ✓ Queried arXiv (42 results)\n  ✓ Queried ACM DL (18 results)\n  ✓ Queried IEEE Xplore (25 results)\n  ✓ Queried Semantic Scholar (67 results)\n  ✓ Deduplicated and ranked\n  ✓ Top 10 results selected\n\n  Duration: 15s\n  Status: COMPLETE\n\nOutput:\n  10 papers identified\n  Saved to: .aiwg\u002Fresearch\u002Fsearch-cache\u002Fresults-20260203-143000.yaml\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nHUMAN GATE: Paper Selection\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nTop 10 Results:\n\n 1. [✓] AutoGen: Enabling Next-Gen LLM Applications (Wu et al., 2023)\n    Relevance: 0.95, Citations: 234, DOI: 10.48550\u002FarXiv.2308.08155\n\n 2. [✓] The Landscape of Emerging AI Agent Architectures (Wang et al., 2024)\n    Relevance: 0.89, Citations: 89, DOI: 10.48550\u002FarXiv.2404.11584\n\n 3. [ ] MetaGPT: Meta Programming for Multi-Agent Systems (Hong et al., 2023)\n    Relevance: 0.87, Citations: 156, DOI: 10.48550\u002FarXiv.2308.00352\n    Note: Already in corpus as REF-013\n\n 4. [✓] Agent Laboratory: Using LLM Agents as Research Assistants (Schmidgall et al., 2024)\n    Relevance: 0.85, Citations: 45, arXiv:2404.11587\n\n... (6 more)\n\nSelect papers to acquire [1,2,4 or 'all']: 1,2,4\n\nSelected: 3 papers\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nWorkflow Progress: [████████░░] Stage 2\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 2: Acquisition (agent: research-acquisition-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n\n  Paper 1\u002F3: AutoGen (10.48550\u002FarXiv.2308.08155)\n    ✓ Downloaded PDF (2.4 MB)\n    ✓ Metadata extracted\n    ✓ Assigned REF-022\n    ✓ Finding document created\n\n  Paper 2\u002F3: Emerging AI Agent Architectures (10.48550\u002FarXiv.2404.11584)\n    ✓ Downloaded PDF (3.1 MB)\n    ✓ Metadata extracted\n    ✓ Assigned REF-075\n    ✓ Finding document created\n\n  Paper 3\u002F3: Agent Laboratory (arXiv:2404.11587)\n    ✓ Downloaded PDF (1.8 MB)\n    ✓ Metadata extracted\n    ✓ Assigned REF-076\n    ✓ Finding document created\n\n  Duration: 42s\n  Status: COMPLETE\n\nOutput:\n  3 papers acquired\n  REF-022, REF-075, REF-076\n  Total size: 7.3 MB\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nWorkflow Progress: [████████████░░] Stage 3\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 3: Documentation (agent: documentation-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n\n  REF-022: AutoGen\n    ✓ PDF parsed (27 pages)\n    ✓ 4 key findings extracted\n    ✓ AIWG relevance assessed (HIGH)\n    ✓ Literature notes created\n    ✓ Finding document populated (1,847 words)\n\n  REF-075: Emerging AI Agent Architectures\n    ✓ PDF parsed (18 pages)\n    ✓ 5 key findings extracted\n    ✓ AIWG relevance assessed (HIGH)\n    ✓ Literature notes created\n    ✓ Finding document populated (2,103 words)\n\n  REF-076: Agent Laboratory\n    ✓ PDF parsed (12 pages)\n    ✓ 3 key findings extracted\n    ✓ AIWG relevance assessed (MEDIUM)\n    ✓ Literature notes created\n    ✓ Finding document populated (1,524 words)\n\n  Duration: 3m 15s\n  Status: COMPLETE\n\nOutput:\n  3 finding documents completed\n  3 literature notes created\n  Total: 5,474 words of documentation\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nWorkflow Progress: [█████████████░] Stage 4\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 4: Quality Assessment (agent: quality-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n\n  REF-022: AutoGen\n    ✓ Baseline: MODERATE (conference paper)\n    ✓ Downgrade: -1 (imprecision)\n    ✓ Final GRADE: LOW\n    ✓ Assessment saved\n\n  REF-075: Emerging AI Agent Architectures\n    ✓ Baseline: VERY LOW (preprint, not peer-reviewed)\n    ✓ No upgrades\u002Fdowngrades\n    ✓ Final GRADE: VERY LOW\n    ✓ Assessment saved\n\n  REF-076: Agent Laboratory\n    ✓ Baseline: MODERATE (preprint, high-quality)\n    ✓ Upgrade: +1 (large effect)\n    ✓ Final GRADE: MODERATE\n    ✓ Assessment saved\n\n  Duration: 45s\n  Status: COMPLETE\n\nOutput:\n  3 quality assessments completed\n  GRADE levels: LOW, VERY LOW, MODERATE\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nHUMAN GATE: Quality Approval\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nQuality assessments complete. Review GRADE levels:\n\n  REF-022: LOW (conference paper with limited evaluation)\n  REF-075: VERY LOW (preprint, not peer-reviewed)\n  REF-076: MODERATE (high-quality preprint with strong findings)\n\nApprove quality levels? [Y\u002Fn]: Y\n\nApproved. Proceeding to archival.\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nWorkflow Progress: [██████████████] Stage 5\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 5: Archival (agent: archival-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n\n  REF-022: AutoGen\n    ✓ BagIt package created (2.5 MB)\n    ✓ Checksums verified\n    ✓ Registered in archival index\n\n  REF-075: Emerging AI Agent Architectures\n    ✓ BagIt package created (3.2 MB)\n    ✓ Checksums verified\n    ✓ Registered in archival index\n\n  REF-076: Agent Laboratory\n    ✓ BagIt package created (1.9 MB)\n    ✓ Checksums verified\n    ✓ Registered in archival index\n\n  Duration: 28s\n  Status: COMPLETE\n\nOutput:\n  3 archival packages created\n  Total archived size: 7.6 MB\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nWorkflow Complete!\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nSummary:\n  Workflow: discovery-to-corpus\n  Duration: 5m 25s\n  Papers processed: 3\n  Success rate: 100%\n\nArtifacts Created:\n  - 3 PDFs (.aiwg\u002Fresearch\u002Fsources\u002F)\n  - 3 finding documents (.aiwg\u002Fresearch\u002Ffindings\u002F)\n  - 3 literature notes (.aiwg\u002Fresearch\u002Fliterature-notes\u002F)\n  - 3 quality assessments (.aiwg\u002Fresearch\u002Fquality-assessments\u002F)\n  - 3 archival packages (.aiwg\u002Fresearch\u002Farchives\u002F)\n\nResource Usage:\n  Tokens consumed: 45,230\n  API calls: 27\n  Storage used: 7.6 MB\n\nNext Steps:\n  - Review findings: \u002Fresearch-document REF-022 REF-075 REF-076\n  - Generate citations: \u002Fresearch-cite REF-022\n  - Check corpus health: \u002Fresearch-status\n\nWorkflow log: .aiwg\u002Fresearch\u002Fworkflows\u002Fwf-20260203-143000.log\n```\n\n## Workflow State\n\nAll workflows track state for resumption:\n\n```yaml\n# .aiwg\u002Fresearch\u002Fworkflows\u002Fwf-20260203-143000-state.yaml\nworkflow_id: wf-20260203-143000\nworkflow_name: discovery-to-corpus\nstatus: complete\nstarted_at: \"2026-02-03T14:30:00Z\"\ncompleted_at: \"2026-02-03T14:35:25Z\"\n\nstages:\n  - name: discovery\n    status: complete\n    started_at: \"2026-02-03T14:30:00Z\"\n    completed_at: \"2026-02-03T14:30:15Z\"\n    output:\n      papers: 10\n      selected: [1, 2, 4]\n\n  - name: acquisition\n    status: complete\n    started_at: \"2026-02-03T14:30:20Z\"\n    completed_at: \"2026-02-03T14:31:02Z\"\n    output:\n      acquired: [REF-022, REF-075, REF-076]\n\n  ... (stages 3-5)\n\nmetrics:\n  duration_seconds: 325\n  tokens_consumed: 45230\n  api_calls: 27\n  success_rate: 1.0\n```\n\n## Custom Workflows\n\nDefine custom workflows in YAML:\n\n```yaml\n# custom-workflow.yaml\nname: focused-acquisition\ndescription: Acquire and document specific papers\nstages:\n  - name: acquisition\n    agent: research-acquisition-agent\n    inputs:\n      - doi_list\n\n  - name: documentation\n    agent: documentation-agent\n    inputs:\n      - from: acquisition.acquired\n\n  - name: quality\n    agent: quality-agent\n    inputs:\n      - from: acquisition.acquired\n\ngates:\n  - stage: quality\n    type: approval\n    message: \"Review quality assessments\"\n```\n\nExecute:\n```bash\n\u002Fresearch-workflow custom-workflow.yaml --input '{\"doi_list\": [\"10.1234\u002Fexample\"]}'\n```\n\n## References\n\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Fframeworks\u002Fresearch-complete\u002Fagents\u002Fworkflow-agent.md - Workflow Agent\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Fframeworks\u002Fresearch-complete\u002Fworkflows\u002F - Workflow definitions\n- @$AIWG_ROOT\u002Fsrc\u002Fresearch\u002Fservices\u002Fworkflow-service.ts - Workflow orchestration\n- @.aiwg\u002Fresearch\u002Fworkflows\u002F - Workflow state and logs\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Fframeworks\u002Fsdlc-complete\u002Frules\u002Fhitl-gates.md - Human gate patterns\n","# Research Workflow Command\n\nEvery acquisition, induction, and synthesis dispatch uses the shared [context engineering contract](..\u002F..\u002F..\u002F..\u002F..\u002F..\u002Fdocs\u002Fcontext-engineering-contract.md) and [verification contract](..\u002F..\u002F..\u002F..\u002F..\u002F..\u002Fdocs\u002Fverification-contracts.md). Synthesis completion requires citation resolution, GRADE-calibrated claims, and evidence\u002Finference\u002Frecommendation separation.\n\nExecute complete multi-stage research workflows from discovery through archival.\n\n## Instructions\n\nWhen invoked, orchestrate multi-agent research workflows:\n\n1. **Load Workflow Definition**\n   - Identify workflow by name or load custom workflow YAML\n   - Parse stages, agents, dependencies\n   - Validate workflow structure\n\n2. **Execute Stages Sequentially**\n   - Each stage invokes specific agents\n   - Pass outputs from one stage to next\n   - Handle stage failures and retries\n   - Track progress and status\n\n3. **Monitor Execution**\n   - Display progress indicators\n   - Log all agent invocations\n   - Capture intermediate outputs\n   - Track resource usage (tokens, time)\n\n4. **Handle Gates**\n   - Pause for human approval at designated gates\n   - Present artifacts for review\n   - Collect feedback and decisions\n   - Resume or abort based on input\n\n5. **Generate Report**\n   - Summarize workflow execution\n   - Report outcomes for each stage\n   - Calculate quality metrics\n   - Archive workflow state\n\n## Built-in Workflows\n\n| Workflow | Stages | Description |\n|----------|--------|-------------|\n| `discovery-to-corpus` | 5 | Full pipeline from search to documented findings |\n| `paper-acquisition` | 3 | Download, extract metadata, create finding document |\n| `quality-assessment` | 4 | GRADE assessment with citation validation |\n| `corpus-maintenance` | 6 | Periodic corpus health checks and updates |\n| `synthesis-report` | 4 | Generate synthesis report from topic cluster |\n| `citation-audit` | 3 | Validate all citations across corpus |\n\n## Arguments\n\n- `[workflow-name]` - Workflow to execute (required)\n- `--input [yaml-file]` - Input parameters for workflow\n- `--stage [n]` - Start from specific stage (default: 1)\n- `--pause-at [stage]` - Pause after specific stage\n- `--interactive` - Prompt for confirmation at each stage\n- `--dry-run` - Preview workflow without execution\n- `--resume [workflow-id]` - Resume previously interrupted workflow\n\n## Workflow Definitions\n\n### discovery-to-corpus\n\nComplete pipeline from literature search to documented findings:\n\n**Stages:**\n\n1. **Discovery** (agent: discovery-agent)\n   - Search academic databases for query\n   - Rank and filter results\n   - Present top candidates\n\n2. **Acquisition** (agent: research-acquisition-agent)\n   - Download selected papers\n   - Extract metadata\n   - Generate frontmatter\n   - Create finding documents\n\n3. **Documentation** (agent: documentation-agent)\n   - Parse PDFs\n   - Extract key findings\n   - Assess AIWG relevance\n   - Generate literature notes\n\n4. **Quality Assessment** (agent: quality-agent)\n   - Apply GRADE framework\n   - Calculate quality level\n   - Generate assessment reports\n   - Update frontmatter\n\n5. **Archival** (agent: archival-agent)\n   - Create BagIt packages\n   - Update fixity manifest\n   - Register in archival index\n\n**Human Gates:**\n- After Discovery: Select papers to acquire\n- After Quality Assessment: Approve quality levels\n\n### paper-acquisition\n\nStreamlined acquisition workflow:\n\n**Stages:**\n\n1. **Download** (agent: research-acquisition-agent)\n   - Fetch PDF from source\n   - Verify file integrity\n\n2. **Metadata Extraction** (agent: research-acquisition-agent)\n   - Parse PDF metadata\n   - Enrich via CrossRef\u002FSemantic Scholar\n   - Assign REF-XXX identifier\n\n3. **Document Creation** (agent: documentation-agent)\n   - Generate finding document from template\n   - Populate frontmatter\n   - Add placeholder sections\n\n### quality-assessment\n\nComprehensive quality assessment workflow:\n\n**Stages:**\n\n1. **GRADE Assessment** (agent: quality-agent)\n   - Determine baseline quality\n   - Apply downgrade\u002Fupgrade factors\n   - Calculate final GRADE level\n\n2. **Hedging Analysis** (agent: quality-agent)\n   - Generate appropriate hedging language\n   - Document forbidden phrases\n   - Create citation templates\n\n3. **Citation Validation** (agent: citation-agent)\n   - Scan corpus for citations of this source\n   - Check hedging compliance\n   - Generate remediation suggestions\n\n4. **Report Generation** (agent: quality-agent)\n   - Create assessment report\n   - Update frontmatter\n   - Save assessment YAML\n\n## Examples\n\n```bash\n# Execute full discovery-to-corpus workflow\n\u002Fresearch-workflow discovery-to-corpus --input discovery-params.yaml\n\n# Acquire specific paper\n\u002Fresearch-workflow paper-acquisition --input '{\"doi\": \"10.48550\u002FarXiv.2308.08155\"}'\n\n# Run quality assessment\n\u002Fresearch-workflow quality-assessment --input '{\"ref_id\": \"REF-022\"}'\n\n# Interactive mode with pauses\n\u002Fresearch-workflow discovery-to-corpus --interactive\n\n# Dry run to preview\n\u002Fresearch-workflow corpus-maintenance --dry-run\n\n# Resume interrupted workflow\n\u002Fresearch-workflow resume wf-20260203-123456\n```\n\n## Expected Output\n\n```\nExecuting Workflow: discovery-to-corpus\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nInput Parameters:\n  Query: \"agentic workflows for software development\"\n  Max results: 10\n  Year from: 2020\n\nWorkflow Progress: [████░░░░░░] Stage 1\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 1: Discovery (agent: discovery-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n  ✓ Queried arXiv (42 results)\n  ✓ Queried ACM DL (18 results)\n  ✓ Queried IEEE Xplore (25 results)\n  ✓ Queried Semantic Scholar (67 results)\n  ✓ Deduplicated and ranked\n  ✓ Top 10 results selected\n\n  Duration: 15s\n  Status: COMPLETE\n\nOutput:\n  10 papers identified\n  Saved to: .aiwg\u002Fresearch\u002Fsearch-cache\u002Fresults-20260203-143000.yaml\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nHUMAN GATE: Paper Selection\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nTop 10 Results:\n\n 1. [✓] AutoGen: Enabling Next-Gen LLM Applications (Wu et al., 2023)\n    Relevance: 0.95, Citations: 234, DOI: 10.48550\u002FarXiv.2308.08155\n\n 2. [✓] The Landscape of Emerging AI Agent Architectures (Wang et al., 2024)\n    Relevance: 0.89, Citations: 89, DOI: 10.48550\u002FarXiv.2404.11584\n\n 3. [ ] MetaGPT: Meta Programming for Multi-Agent Systems (Hong et al., 2023)\n    Relevance: 0.87, Citations: 156, DOI: 10.48550\u002FarXiv.2308.00352\n    Note: Already in corpus as REF-013\n\n 4. [✓] Agent Laboratory: Using LLM Agents as Research Assistants (Schmidgall et al., 2024)\n    Relevance: 0.85, Citations: 45, arXiv:2404.11587\n\n... (6 more)\n\nSelect papers to acquire [1,2,4 or 'all']: 1,2,4\n\nSelected: 3 papers\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nWorkflow Progress: [████████░░] Stage 2\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 2: Acquisition (agent: research-acquisition-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n\n  Paper 1\u002F3: AutoGen (10.48550\u002FarXiv.2308.08155)\n    ✓ Downloaded PDF (2.4 MB)\n    ✓ Metadata extracted\n    ✓ Assigned REF-022\n    ✓ Finding document created\n\n  Paper 2\u002F3: Emerging AI Agent Architectures (10.48550\u002FarXiv.2404.11584)\n    ✓ Downloaded PDF (3.1 MB)\n    ✓ Metadata extracted\n    ✓ Assigned REF-075\n    ✓ Finding document created\n\n  Paper 3\u002F3: Agent Laboratory (arXiv:2404.11587)\n    ✓ Downloaded PDF (1.8 MB)\n    ✓ Metadata extracted\n    ✓ Assigned REF-076\n    ✓ Finding document created\n\n  Duration: 42s\n  Status: COMPLETE\n\nOutput:\n  3 papers acquired\n  REF-022, REF-075, REF-076\n  Total size: 7.3 MB\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nWorkflow Progress: [████████████░░] Stage 3\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 3: Documentation (agent: documentation-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n\n  REF-022: AutoGen\n    ✓ PDF parsed (27 pages)\n    ✓ 4 key findings extracted\n    ✓ AIWG relevance assessed (HIGH)\n    ✓ Literature notes created\n    ✓ Finding document populated (1,847 words)\n\n  REF-075: Emerging AI Agent Architectures\n    ✓ PDF parsed (18 pages)\n    ✓ 5 key findings extracted\n    ✓ AIWG relevance assessed (HIGH)\n    ✓ Literature notes created\n    ✓ Finding document populated (2,103 words)\n\n  REF-076: Agent Laboratory\n    ✓ PDF parsed (12 pages)\n    ✓ 3 key findings extracted\n    ✓ AIWG relevance assessed (MEDIUM)\n    ✓ Literature notes created\n    ✓ Finding document populated (1,524 words)\n\n  Duration: 3m 15s\n  Status: COMPLETE\n\nOutput:\n  3 finding documents completed\n  3 literature notes created\n  Total: 5,474 words of documentation\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nWorkflow Progress: [█████████████░] Stage 4\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 4: Quality Assessment (agent: quality-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n\n  REF-022: AutoGen\n    ✓ Baseline: MODERATE (conference paper)\n    ✓ Downgrade: -1 (imprecision)\n    ✓ Final GRADE: LOW\n    ✓ Assessment saved\n\n  REF-075: Emerging AI Agent Architectures\n    ✓ Baseline: VERY LOW (preprint, not peer-reviewed)\n    ✓ No upgrades\u002Fdowngrades\n    ✓ Final GRADE: VERY LOW\n    ✓ Assessment saved\n\n  REF-076: Agent Laboratory\n    ✓ Baseline: MODERATE (preprint, high-quality)\n    ✓ Upgrade: +1 (large effect)\n    ✓ Final GRADE: MODERATE\n    ✓ Assessment saved\n\n  Duration: 45s\n  Status: COMPLETE\n\nOutput:\n  3 quality assessments completed\n  GRADE levels: LOW, VERY LOW, MODERATE\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nHUMAN GATE: Quality Approval\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nQuality assessments complete. Review GRADE levels:\n\n  REF-022: LOW (conference paper with limited evaluation)\n  REF-075: VERY LOW (preprint, not peer-reviewed)\n  REF-076: MODERATE (high-quality preprint with strong findings)\n\nApprove quality levels? [Y\u002Fn]: Y\n\nApproved. Proceeding to archival.\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nWorkflow Progress: [██████████████] Stage 5\u002F5\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nStage 5: Archival (agent: archival-agent)\n─────────────────────────────────────────────────────────────────────\n  Status: Running...\n\n  REF-022: AutoGen\n    ✓ BagIt package created (2.5 MB)\n    ✓ Checksums verified\n    ✓ Registered in archival index\n\n  REF-075: Emerging AI Agent Architectures\n    ✓ BagIt package created (3.2 MB)\n    ✓ Checksums verified\n    ✓ Registered in archival index\n\n  REF-076: Agent Laboratory\n    ✓ BagIt package created (1.9 MB)\n    ✓ Checksums verified\n    ✓ Registered in archival index\n\n  Duration: 28s\n  Status: COMPLETE\n\nOutput:\n  3 archival packages created\n  Total archived size: 7.6 MB\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\nWorkflow Complete!\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\nSummary:\n  Workflow: discovery-to-corpus\n  Duration: 5m 25s\n  Papers processed: 3\n  Success rate: 100%\n\nArtifacts Created:\n  - 3 PDFs (.aiwg\u002Fresearch\u002Fsources\u002F)\n  - 3 finding documents (.aiwg\u002Fresearch\u002Ffindings\u002F)\n  - 3 literature notes (.aiwg\u002Fresearch\u002Fliterature-notes\u002F)\n  - 3 quality assessments (.aiwg\u002Fresearch\u002Fquality-assessments\u002F)\n  - 3 archival packages (.aiwg\u002Fresearch\u002Farchives\u002F)\n\nResource Usage:\n  Tokens consumed: 45,230\n  API calls: 27\n  Storage used: 7.6 MB\n\nNext Steps:\n  - Review findings: \u002Fresearch-document REF-022 REF-075 REF-076\n  - Generate citations: \u002Fresearch-cite REF-022\n  - Check corpus health: \u002Fresearch-status\n\nWorkflow log: .aiwg\u002Fresearch\u002Fworkflows\u002Fwf-20260203-143000.log\n```\n\n## Workflow State\n\nAll workflows track state for resumption:\n\n```yaml\n# .aiwg\u002Fresearch\u002Fworkflows\u002Fwf-20260203-143000-state.yaml\nworkflow_id: wf-20260203-143000\nworkflow_name: discovery-to-corpus\nstatus: complete\nstarted_at: \"2026-02-03T14:30:00Z\"\ncompleted_at: \"2026-02-03T14:35:25Z\"\n\nstages:\n  - name: discovery\n    status: complete\n    started_at: \"2026-02-03T14:30:00Z\"\n    completed_at: \"2026-02-03T14:30:15Z\"\n    output:\n      papers: 10\n      selected: [1, 2, 4]\n\n  - name: acquisition\n    status: complete\n    started_at: \"2026-02-03T14:30:20Z\"\n    completed_at: \"2026-02-03T14:31:02Z\"\n    output:\n      acquired: [REF-022, REF-075, REF-076]\n\n  ... (stages 3-5)\n\nmetrics:\n  duration_seconds: 325\n  tokens_consumed: 45230\n  api_calls: 27\n  success_rate: 1.0\n```\n\n## Custom Workflows\n\nDefine custom workflows in YAML:\n\n```yaml\n# custom-workflow.yaml\nname: focused-acquisition\ndescription: Acquire and document specific papers\nstages:\n  - name: acquisition\n    agent: research-acquisition-agent\n    inputs:\n      - doi_list\n\n  - name: documentation\n    agent: documentation-agent\n    inputs:\n      - from: acquisition.acquired\n\n  - name: quality\n    agent: quality-agent\n    inputs:\n      - from: acquisition.acquired\n\ngates:\n  - stage: quality\n    type: approval\n    message: \"Review quality assessments\"\n```\n\nExecute:\n```bash\n\u002Fresearch-workflow custom-workflow.yaml --input '{\"doi_list\": [\"10.1234\u002Fexample\"]}'\n```\n\n## References\n\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Fframeworks\u002Fresearch-complete\u002Fagents\u002Fworkflow-agent.md - Workflow Agent\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Fframeworks\u002Fresearch-complete\u002Fworkflows\u002F - Workflow definitions\n- @$AIWG_ROOT\u002Fsrc\u002Fresearch\u002Fservices\u002Fworkflow-service.ts - Workflow orchestration\n- @.aiwg\u002Fresearch\u002Fworkflows\u002F - Workflow state and logs\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Fframeworks\u002Fsdlc-complete\u002Frules\u002Fhitl-gates.md - Human gate patterns\n",{"name":5,"triggers":49,"namespace":53,"platforms":54,"commandHint":56,"description":7},[50,51,52,29],"help with research papers","help me build a cited research corpus","start a research workflow or citation plan","aiwg",[55],"all",{"category":57,"argumentHint":58},"research-workflows","[workflow-name] [--input parameters] [--stage n]","main","6754967c344ccb087a8de06972f55a41067b291b","agentic\u002Fcode\u002Fframeworks\u002Fresearch-complete\u002Fskills\u002Fresearch-workflow",36,"https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg","https:\u002F\u002Faiwg.io","https:\u002F\u002Fgithub.com\u002Fjmagly","https:\u002F\u002Fintegrolabs.io","ml\u002Fai, industrial automation, safety\u002Fsecurity, ssi\r\n\r\nmagly.net\r\nintegrolabs.io\r\naiwg.io\r\nfortemi.com\r\npagenary.com\r\nbytecask.com","2026-11-14 02:10:01",{"source":8,"skillName":5,"commands":70,"prompt":80,"zipAvailable":81},[71,74,77],{"tool":72,"command":73},"npx","npx skills add jmagly\u002Faiwg --skill research-workflow",{"tool":75,"command":76},"bunx","bunx skills add jmagly\u002Faiwg --skill research-workflow",{"tool":78,"command":79},"pnpm","pnpm dlx skills add jmagly\u002Faiwg --skill research-workflow","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fframeworks\u002Fresearch-complete\u002Fskills\u002Fresearch-workflow\u002FSKILL.md",true,[],[84,111,135,141,148,162],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":91,"publishedAt":44},"01M70QA0BSYJ49EES9V4RE8QH1","host-hardening","Read-only system hardening baseline and plan-first host hardening recommendations for SSH, Docker bind addresses, firewall persistence, datastores, exporters and audit retention.","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fplugins\u002Fsecurity-engineering\u002Fskills\u002Fhost-hardening\u002FSKILL.md",[90],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[92,93,94,95,96,97,98,40,99,100,101,102,103,104,105,106,107,108,109,110],"host","host hardening","hardening","read","re","recommendations","remediation","port","verification","Bash","access","only","retention","audit","never","firewall","bind","secrets","request",{"skillNo":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":121,"publishedAt":44},"01M70PNT2Z4HMZPRBTAF6073NN","exposure-analysis","Read-only firewall analysis and effective host exposure reconstruction for Docker published ports, UFW bypass, stale rules, CDN scope, IPv6 asymmetry and external probe allowlists.","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fplugins\u002Fforensics\u002Fskills\u002Fexposure-analysis\u002FSKILL.md",[117,120],{"categoryNo":118,"categoryName":119,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000011","运维与部署",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[122,123,124,95,125,99,126,127,101,128,92,129,103,130,131,106,107,132,133,134],"exposure analysis","analysis","unknown","credentials","collection","ufw","ports","internet","local","scope","reconstruction","operator","write",{"skillNo":136,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":140,"publishedAt":44},"01M70P8QATMPKA78E4FT5D2TNA","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fframeworks\u002Fsecurity-engineering\u002Fskills\u002Fhost-hardening\u002FSKILL.md",[139],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[94,92,93,102,95,110,98,97,105,40,99,100,101,109,103,104,96,106,107,108],{"skillNo":142,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":143,"categories":144,"keywords":147,"publishedAt":44},"01M70NMPVJ5H0JH131P60FW4R8","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fframeworks\u002Fforensics-complete\u002Fskills\u002Fexposure-analysis\u002FSKILL.md",[145,146],{"categoryNo":118,"categoryName":119,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[123,122,106,125,99,126,101,124,92,129,103,130,131,95,107,128,127,132,133,134],{"skillNo":149,"name":150,"nameZh":6,"description":151,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":152,"categories":153,"keywords":155,"publishedAt":44},"01M6CB4DDR0B4NNCFK7ZK7449H","fixture-skill","Exercise rule delivery.","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Frule-delivery\u002Fskills\u002Ffixture-skill\u002FSKILL.md",[154],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[156,157,158,159,160,161],"fixture skill","fixture","rule","delivery","true","exercise",{"skillNo":163,"name":164,"nameZh":6,"description":165,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":166,"categories":167,"keywords":169,"publishedAt":44},"01M6CB4AVK1W76MFMPP0JP32CJ","pi-local","Pi-local duplicate-precedence fixture.","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Fproviders\u002Fpi\u002Fproject\u002F.pi\u002Fskills\u002Fpi-local\u002FSKILL.md",[168],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[170,171,130,172,157],"pi","pi local","duplicate"]