community-research-insight

name	community-research-insight
description	Extract structured insight briefs from community research transcripts or notes. Produces pain points, stakeholder needs, opportunity maps, risks, and follow-up questions. Requires human review before publication.
kind	meta
meta_priority	60
always	false
final_text_mode	step:final_response
triggers	["community research","research insight","community insight","transcript analysis","社区研究","研究洞察","访谈分析","社区洞察"]
provenance	{"origin":"openclaw.net","license":"MIT"}
composition	{"steps":[{"id":"collect","kind":"user_input","with":{"prompt":"Paste the community research transcript or notes, plus any project\ncontext and target audience. Include only source material you want the\ninsight brief to analyze.\n"},"clarify":{"mode":"chat","extract_natural_language":true,"cancel_words":["cancel","取消","stop","abort"],"timeout_seconds":1800}},{"id":"analyze","kind":"llm_chat","depends_on":["collect"],"on_failure":"analyze_fallback","output_contract":{"format":"json","required_properties":["key_pain_points","stakeholder_needs","opportunities"]},"with":{"system_prompt":"You are a community research analyst. Extract grounded themes from\nthe provided transcript or notes. Do not invent quotes or attribute\nviews to named people unless the source explicitly does. Return ONLY\none JSON object and do not use markdown fences.\n","input":"Analyze the following community research material and return ONLY a\nsingle JSON object.\n\nTranscript or notes:\n{{ outputs.collect | truncate(50000) }}\n\nReturn this JSON structure:\n{\n \"key_pain_points\": [\"...\"],\n \"stakeholder_needs\": [\"...\"],\n \"opportunities\": [\"...\"],\n \"evidence_summary\": \"...\",\n \"gaps\": [\"...\"]\n}\n\nRules:\n- Every pain point must be grounded in the source.\n- Separate evidence from inference.\n- Flag missing context under the \"gaps\" key.\n- Do not invent quotes, names, dates, or statistics.\n- Do not recommend replacing community engagement with automation.\n"}},{"id":"analyze_fallback","kind":"llm_chat","output_contract":{"format":"json","required_properties":["key_pain_points","stakeholder_needs","opportunities"]},"with":{"system_prompt":"You are a community research analyst on a fallback path. Produce a\ngrounded, best-effort JSON analysis from the available source material.\nReturn ONLY one JSON object. Do not use markdown fences.\n","input":"Analyze the following research material and return structured JSON:\n\n{{ outputs.collect | truncate(30000) }}\n\nReturn this JSON structure:\n{\n \"key_pain_points\": [\"...\"],\n \"stakeholder_needs\": [\"...\"],\n \"opportunities\": [\"...\"],\n \"evidence_summary\": \"...\",\n \"gaps\": [\"...\"]\n}\n"}},{"id":"draft","kind":"llm_chat","depends_on":["analyze"],"output_contract":{"format":"json","required_properties":["executive_summary","key_pain_points","stakeholder_needs","opportunity_map","risks_and_cautions","follow_up_questions"]},"with":{"system_prompt":"You are drafting a community research insight brief for human review.\nReturn ONLY a single JSON object. Do NOT wrap it in ```json fences.\nDo NOT include ANY text before or after the JSON. Your entire response\nmust be parseable as a JSON object starting with { and ending with }.\nEvery key claim must be grounded in the analysis results provided.\nDo not invent quotes, names, dates, or statistics not present in the\nsource.\n","input":"Draft a complete insight brief from the analysis below.\nOutput ONLY the JSON — no markdown, no explanation, no code fences.\n\nAnalysis results:\n{{ outputs.analyze | truncate(8000) }}\n\nProject context:\n{{ outputs.collect | truncate(2000) }}\n\nReturn this exact JSON structure (no markdown fences):\n{\n \"executive_summary\": \"<2-4 sentences>\",\n \"key_pain_points\": [\n {\"issue\": \"...\", \"evidence\": \"...\", \"severity\": \"high|medium|low\"}\n ],\n \"stakeholder_needs\": [\n {\"group\": \"...\", \"need\": \"...\", \"urgency\": \"high|medium|low\"}\n ],\n \"opportunity_map\": [\n {\"opportunity\": \"...\", \"feasibility\": \"high|medium|low\", \"impact\": \"high|medium|low\"}\n ],\n \"risks_and_cautions\": [\n {\"risk\": \"...\", \"likelihood\": \"high|medium|low\", \"mitigation\": \"...\"}\n ],\n \"follow_up_questions\": [\"...\"],\n \"missing_information\": [\"...\"],\n \"attribution_note\": \"<explicit statement about what is evidence vs inference>\"\n}\n"}},{"id":"validate","kind":"llm_chat","depends_on":["draft"],"output_choices":["PASS","REVISE"],"route":[{"when":"{{ outputs.validate == 'PASS' }}","to":"preview"},{"to":"validation_revise"}],"with":{"system_prompt":"You are a strict grounding validator. Return exactly PASS or REVISE.\n","input":"Classify whether this draft brief is ready for human preview.\nReturn PASS only if every checklist item is satisfied. Return REVISE\nif any claim lacks grounding, overstates automation, omits risks, or\ninvents quotes, names, dates, or statistics.\n\nDraft brief:\n{{ outputs.draft | truncate(8000) }}\n\nOriginal analysis:\n{{ outputs.analyze | truncate(5000) }}\n\nValidation checklist:\n- Every key claim is grounded in the transcript or provided context.\n- Recommendations are framed as support tools, not replacements for relationships.\n- Missing information is listed separately.\n- Risks are included alongside opportunities.\n- No invented quotes, names, dates, or statistics.\n"}},{"id":"validation_revise","kind":"llm_chat","depends_on":["validate"],"with":{"system_prompt":"You are a grounding validator. The draft brief did not pass validation.\nExplain the revision requirement clearly and do not publish the brief.\n","input":"The draft research insight brief needs revision before preview.\n\nDraft brief:\n{{ outputs.draft | truncate(8000) }}\n\nOriginal analysis:\n{{ outputs.analyze | truncate(5000) }}\n\nReturn a concise Markdown message explaining that the brief is blocked\npending revision and list the likely grounding, risk, or attribution\nissues to address.\n"}},{"id":"preview","kind":"llm_chat","depends_on":["draft","validate"],"with":{"system_prompt":"You are preparing a research insight brief for human review.\nPresent the validated findings in a clean Markdown format.\n","input":"Produce a human-readable preview of the validated insight brief.\n\nValidated draft:\n{{ outputs.draft | truncate(8000) }}\n\nValidation result:\n{{ outputs.validate | truncate(2000) }}\n\nInclude:\n1. Executive Summary\n2. Key Pain Points (with severity)\n3. Stakeholder Needs (with urgency)\n4. Opportunity Map (with feasibility and impact)\n5. Risks and Cautions (with mitigations)\n6. Follow-up Questions\n7. Missing Information\n8. Attribution Note\n\nEnd with:\n---\n**Status**: Awaiting human review.\nReply \"approve\", \"revise\", or \"reject\" to proceed.\n"}},{"id":"review","kind":"user_input","depends_on":["preview"],"with":{"prompt":"{{ outputs.preview | truncate(12000) }}\n"},"clarify":{"mode":"chat","extract_natural_language":true,"cancel_words":["reject","拒绝","abort"],"timeout_seconds":86400}},{"id":"final_response","kind":"llm_chat","depends_on":["preview","review"],"with":{"system_prompt":"You are finalizing a community research insight brief after human\nreview. Respect the reviewer decision exactly.\n","input":"Review decision and feedback:\n{{ outputs.review | truncate(2000) }}\n\n{% if 'approve' in outputs.review %}\nProduce the final approved brief from the preview below.\nMark it as APPROVED and ready for distribution.\n\n{{ outputs.preview | truncate(8000) }}\n\n{% elif 'revise' in outputs.review %}\nRevision requested. Feedback:\n{{ outputs.review | truncate(2000) }}\n\nRe-analyze the source material incorporating this feedback and produce\na revised Markdown brief. Keep all claims grounded in the transcript.\n\nOriginal transcript:\n{{ outputs.collect | truncate(50000) }}\n\n{% else %}\nBrief rejected. No further action.\nReason from reviewer:\n{{ outputs.review | truncate(1000) }}\n{% endif %}\n"}}]}

Community Research Insight Extractor

Extracts pain points, stakeholder needs, risks, and practical technology opportunities from community-engaged research discussions. Produces a structured insight brief for human review before publication.

What It Does

Step	Kind	Purpose
collect	user_input	Collect transcript, context, and audience via chat
analyze	llm_chat	Extract grounded themes as structured JSON
analyze_fallback	llm_chat	Produce best-effort grounded JSON if primary analysis fails
draft	llm_chat	Draft the full 6-section insight brief as structured JSON
validate	llm_chat	Gate preview on PASS vs REVISE grounding validation
validation_revise	llm_chat	Explain why the brief is blocked when validation fails
preview	llm_chat	Render validated findings as human-readable Markdown
review	user_input	Pause for human approve/revise/reject decision
final_response	llm_chat	Produce final output based on review decision

Guardrails

• Never invent quotes, names, dates, or statistics.
• Never attribute views to named people unless present in the source.
• Never recommend replacing community engagement with automation.
• Always separate evidence from inference.
• Always flag missing information rather than filling gaps.
• Always require human review before publication or named attribution.

Fallback

If analyze fails (timeout, provider error, or JSON contract failure), analyze_fallback runs a single-turn llm_chat on the same transcript and must satisfy the same JSON output contract. If validate returns REVISE, the preview path is blocked and validation_revise explains what must be fixed before human review.

Output Contract

The analyze, analyze_fallback, and draft steps enforce OutputContract JSON validation. The draft step requires executive_summary, key_pain_points, stakeholder_needs, opportunity_map, risks_and_cautions, and follow_up_questions.

Safety

Outputs are decision-support drafts for human review. They are not final professional advice in research, policy, or community engagement contexts. Named attribution and external publication require explicit reviewer approval.