brycewang-stanford/Auto-Empirical-Research-Skills / skills/35-bahayonghang-academic-writing-skills/skills/industrial-ai-research
industrial-ai-research
VerifiedIndustrial AI literature research with mandatory intake questions, venue-aware source prioritization, structured report outputs, and survey draft generation. Use when the user needs up-to-date research on predictive maintenance, intelligent scheduling, industrial anomaly detection, smart manufacturing, cyber-physical systems, edge AI for automation, or crossover robotics-for-industry topics. Also trigger for adjacent terms: "digital twin", "industrial IoT", "Industry 4.0", "manufacturing AI", "factory automation", "process optimization", or "survey draft" in industrial contexts.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 12.7 KB |
| agents/openai.yaml | config | 420 B |
| evals/evals.json | config | 6.5 KB |
| examples/industrial-anomaly-detection.md | markdown | 562 B |
| examples/intelligent-scheduling.md | markdown | 521 B |
| examples/predictive-maintenance.md | markdown | 536 B |
| examples/survey-predictive-maintenance.md | markdown | 8.8 KB |
| references/SURVEY_WRITING_GUIDE.md | markdown | 6.3 KB |
| references/modules/SURVEY_EVIDENCE.md | markdown | 3.8 KB |
| references/modules/SURVEY_MERGE.md | markdown | 4.4 KB |
| references/modules/SURVEY_OUTLINE.md | markdown | 4.0 KB |
| references/modules/SURVEY_WRITER.md | markdown | 4.3 KB |
| references/quality-checklist.md | markdown | 1.2 KB |
| references/question-flow.md | markdown | 1.8 KB |
| references/report-modes.md | markdown | 2.9 KB |
| references/source-priority.md | markdown | 2.2 KB |
| references/venue-map.md | markdown | 2.9 KB |
evals/evals.json
View on GitHub{
"skill_name": "industrial-ai-research",
"evals": [
{
"id": 1,
"prompt": "Research recent predictive maintenance work in the last 3 years and return a research-brief that separates peer-reviewed papers from preprints.",
"expected_output": "Ask the mandatory intake questions if they are not already resolved, then produce a research-brief with scope, source buckets, shortlisted papers, evidence-backed synthesis, and explicit preprint labeling.",
"files": [],
"assertions": [
{"type": "regex", "pattern": "(intake|language|deliverable|time window|emphasis)", "description": "intake questions asked"},
{"type": "regex", "pattern": "(preprint|[Pp]reprint|arXiv)", "description": "preprints labeled"},
{"type": "regex", "pattern": "(source bucket|venue|arXiv|IEEE)", "description": "source buckets present"}
]
},
{
"id": 2,
"prompt": "I need the latest Industrial AI papers on intelligent scheduling for job shops. Prioritize the newest work and tell me where the evidence is thin.",
"expected_output": "Prefer recent arXiv plus top automation venues, explain the recency policy, surface thin-evidence areas, and avoid venue-blind mixing in the final report.",
"files": [],
"assertions": [
{"type": "regex", "pattern": "(intake|language|deliverable|time window)", "description": "intake questions asked"},
{"type": "regex", "pattern": "(thin|sparse|limited|gap)", "description": "thin evidence areas surfaced"},
{"type": "regex", "pattern": "(arXiv|IEEE|T-ASE|CASE)", "description": "venue-aware sourcing"}
]
},
{
"id": 3,
"prompt": "Use survey-draft mode for industrial anomaly detection and stop after the YAML outline so I can approve the structure first.",
"expected_output": "Route into the survey-draft workflow, ask the survey-specific follow-up questions if missing, build only the S1 YAML outline, and stop before evidence packs or prose.",
"files": [],
"assertions": [
{"type": "regex", "pattern": "(survey.draft|S1|outline)", "description": "survey-draft mode selected"},
{"type": "regex", "pattern": "(YAML|yaml|taxonomy)", "description": "YAML outline format used"},
{"type": "regex", "pattern": "(approv|checkpoint|stop)", "description": "checkpoint before S2"}
]
},
{
"id": 4,
"prompt": "My topic is hospital triage optimization with clinical ethics constraints. If that is outside Industrial AI scope, redirect me to the closest supported framing instead of pretending it fits.",
"expected_output": "Identify the topic as off-scope, state the boundary clearly, suggest the closest supported Industrial AI framing, and avoid starting a normal synthesis until scope is resolved.",
"files": [],
"assertions": [
{"type": "regex", "pattern": "(out.?of.?scope|outside|boundary|not.*supported)", "description": "off-scope identified"},
{"type": "regex", "pattern": "(closest|suggest|alternative|redirect)", "description": "alternative suggested"},
{"type": "not_contains", "text": "source buckets", "description": "no premature synthesis"}
]
},
{
"id": 5,
"prompt": "Create a literature map comparing deep reinforcement learning methods for robotic manipulation across top robotics venues (ICRA, IROS, RSS, CoRL) from 2022-2025. Show method evolution and benchmark convergence.",
"mode": "literature-map",
"expected_output": "Ask intake questions, produce a literature map with venue-grouped papers, method evolution timeline, and benchmark convergence analysis across ICRA, IROS, RSS, CoRL.",
"files": [],
"assertions": [
{"type": "contains", "text": "ICRA", "description": "venue coverage"},
{"type": "contains", "text": "reinforcement learning", "description": "topic match"},
{"type": "regex", "pattern": "(comparison|evolution|convergence)", "description": "comparative analysis present"}
]
},
{
"id": 6,
"prompt": "Write a venue-ranked survey on federated learning for industrial IoT with explicit contradiction analysis between results from different research groups.",
"mode": "venue-ranked survey",
"expected_output": "Ask intake questions, produce a venue-ranked survey with papers grouped by tier, contradiction analysis between conflicting results, and explicit discussion of methodological disagreements.",
"files": [],
"assertions": [
{"type": "contains", "text": "federated learning", "description": "topic present"},
{"type": "regex", "pattern": "(contradict|conflict|disagree|inconsisten)", "description": "contradiction analysis present"},
{"type": "regex", "pattern": "(Tier[- ]?[1-3]|A\\*|venue)", "description": "venue ranking present"}
]
},
{
"id": 7,
"prompt": "Identify research gaps at the intersection of cyber-physical systems (CPS) and edge AI for smart manufacturing. Focus on real-time safety guarantees and latency-bounded inference.",
"mode": "research-gap memo",
"expected_output": "Ask intake questions, produce a research-gap memo identifying open problems at the CPS-edge AI intersection, focusing on real-time safety and latency constraints.",
"files": [],
"assertions": [
{"type": "contains", "text": "CPS", "description": "CPS topic covered"},
{"type": "regex", "pattern": "(edge|Edge)", "description": "edge AI covered"},
{"type": "regex", "pattern": "(gap|underexplored|open problem|future)", "description": "gap identification present"}
]
},
{
"id": 8,
"prompt": "Summarize the latest advances in Transformer-based defect detection for industrial visual inspection in both Chinese and English, including data augmentation strategies and few-shot solutions.",
"mode": "any",
"expected_output": "Ask intake questions (including confirming bilingual output), produce a bilingual summary covering Transformer-based defect detection, data augmentation, and few-shot approaches.",
"files": [],
"assertions": [
{"type": "contains", "text": "Transformer", "description": "topic match"},
{"type": "regex", "pattern": "(bilingual|Chinese.*English|English.*Chinese)", "description": "bilingual output confirmed"},
{"type": "regex", "pattern": "(few[- ]?shot|data augment)", "description": "key technique covered"},
{"type": "regex", "pattern": "(intake|language|deliverable)", "description": "intake questions asked"}
]
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |