[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6EKXEVFDF9YCZ6WD3RRT4NR":3,"skill-same:01M6EKXEVFDF9YCZ6WD3RRT4NR":70},{"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":18,"publishedAt":29,"linkOnly":30,"markdownContent":31,"markdownBody":32,"frontMatter":33,"refName":47,"commitSha":48,"skillDirectory":49,"forksCount":50,"repositoryHtmlUrl":51,"repositoryHomepageUrl":6,"ownerHtmlUrl":52,"ownerBlogUrl":53,"ownerDescription":54,"updatedAt":55,"install":56,"resources":69},"01M6EKXEVFDF9YCZ6WD3RRT4NR","findings-memory-novelty-cross-run",null,"When findings memory is available, build research questions on prior runs (don't replicate); Critic adds an optional novelty review.","cgpan\u002Fedm-ars-public","cgpan","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F118746117?v=4","01M6EKT5TG36Q3GH340QN8RH08","Chenguang Pan","01M6EKT5YTQVRARGJF0VPA1SY7",5,"MIT","https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public\u002Fblob\u002Fmain\u002Fskills\u002Fmethodology\u002Ffindings-memory-novelty-cross-run\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28],"memory","run","findings memory novelty cross run","review","research","methodology","critic","replicate","build","replication","2026-11-10 10:12:30",false,"---\nname: findings-memory-novelty-cross-run\nlayer: methodology\ndescription: When findings memory is available, build research questions on prior runs (don't replicate); Critic adds an optional novelty review.\ntrigger_keywords:\n  - findings-memory\n  - prior-runs\n  - novelty\n  - replication\napplicable_task_types: []\napplicable_datasets: []\napplicable_stages:\n  - ProblemFormulator\n  - Critic\npriority: 3\nreferences_skills: []\nresources: []\nversion: \"1.0\"\n---\n\n# Findings Memory and Cross-Run Novelty\n\nWhen `## Findings Memory Summary` appears in the user message, the\nProblemFormulator and Critic both treat it as authoritative context about\nwhat prior runs have already produced. The goal is to keep each run\ncontributing new knowledge instead of restating existing findings.\n\n## ProblemFormulator obligations\n\nWhen findings memory is present:\n\n1. Treat the memory as a **catalogue of what has already been studied**.\n   Do not replicate prior research questions verbatim; build on open\n   questions or explore unstudied outcomes.\n2. If prior runs surfaced strong predictors (e.g., \"X1TXMTSCOR appeared\n   frequently\"), you may include them — but the **combination** of\n   outcome + predictor set should be meaningfully novel.\n3. Prefer outcomes listed under `## Studied Outcomes (already\n   investigated in prior runs)` only if the user prompt specifically\n   requests them or no viable alternative exists.\n\nWhen `## Prior Candidate Specs (already generated this session)` is\npresent (multi-candidate generation mode):\n\n1. Your spec MUST differ meaningfully from each prior candidate. Choose a\n   different outcome variable, a different theoretical framing, or a\n   substantially different predictor set.\n2. Do not simply swap one predictor for a near-synonym. Aim for genuine\n   conceptual diversity.\n\n## Critic novelty_review section\n\nWhen findings memory is present, the Critic adds an optional\n`\"novelty_review\"` key to its JSON output (after `revision_instructions`):\n\n```json\n{\n  \"novelty_review\": {\n    \"score\": 7,\n    \"compared_to_prior_runs\": \"This study examines \u003Coutcome> using \u003Cpredictors>. Prior runs focused on \u003Cother-outcome>, so the outcome is novel, but the predictor set overlaps substantially with run_\u003Cid>.\",\n    \"contribution_builds_on_memory\": true\n  }\n}\n```\n\nField semantics:\n\n- `score` (1–10): how novel this run is relative to the accumulated\n  memory.\n- `compared_to_prior_runs`: 1–3 sentences comparing to specific prior\n  runs.\n- `contribution_builds_on_memory`: `true` if the research question\n  extends or responds to open questions or strong predictors from prior\n  runs.\n\nThis section is **optional**. Omit it entirely if no `## Findings Memory\nSummary` is in the user message. Do not include `\"novelty_review\": null`\n— simply omit the key.\n\n## Verification rules (Critic checklist `pf_08`, `pf_09`)\n\nWhen findings memory is enabled (the orchestrator sets a flag when the\nmemory file is non-empty):\n\n1. `pf_08` (minor): if the outcome variable and a substantially similar\n   predictor set have already been studied in a prior run (per\n   `studied_outcomes`), flag as minor unless the research question offers\n   a meaningfully different framing or predictor combination. Use\n   `novelty_review.score` as supporting evidence.\n2. `pf_09` (minor): verify that the research question or\n   `expected_contribution` references open questions or strong predictors\n   from the findings memory when relevant. Aspirational — minor, not\n   blocking.\n\n## Source provenance\n\nCanonical source: `agent_prompts\u002Fproblem_formulator.yaml` §\"Using\nFindings Memory\" + §\"Generating Diverse Candidates\".\n\nMerged content from:\n- `agent_prompts\u002Fcritic.yaml` §\"Optional: Novelty Review Against Prior\n  Runs\"\n- `data_registry\u002Fevaluation_rubrics\u002Fmethodological_checklist.yaml` items\n  `pf_08`, `pf_09`\n","# Findings Memory and Cross-Run Novelty\n\nWhen `## Findings Memory Summary` appears in the user message, the\nProblemFormulator and Critic both treat it as authoritative context about\nwhat prior runs have already produced. The goal is to keep each run\ncontributing new knowledge instead of restating existing findings.\n\n## ProblemFormulator obligations\n\nWhen findings memory is present:\n\n1. Treat the memory as a **catalogue of what has already been studied**.\n   Do not replicate prior research questions verbatim; build on open\n   questions or explore unstudied outcomes.\n2. If prior runs surfaced strong predictors (e.g., \"X1TXMTSCOR appeared\n   frequently\"), you may include them — but the **combination** of\n   outcome + predictor set should be meaningfully novel.\n3. Prefer outcomes listed under `## Studied Outcomes (already\n   investigated in prior runs)` only if the user prompt specifically\n   requests them or no viable alternative exists.\n\nWhen `## Prior Candidate Specs (already generated this session)` is\npresent (multi-candidate generation mode):\n\n1. Your spec MUST differ meaningfully from each prior candidate. Choose a\n   different outcome variable, a different theoretical framing, or a\n   substantially different predictor set.\n2. Do not simply swap one predictor for a near-synonym. Aim for genuine\n   conceptual diversity.\n\n## Critic novelty_review section\n\nWhen findings memory is present, the Critic adds an optional\n`\"novelty_review\"` key to its JSON output (after `revision_instructions`):\n\n```json\n{\n  \"novelty_review\": {\n    \"score\": 7,\n    \"compared_to_prior_runs\": \"This study examines \u003Coutcome> using \u003Cpredictors>. Prior runs focused on \u003Cother-outcome>, so the outcome is novel, but the predictor set overlaps substantially with run_\u003Cid>.\",\n    \"contribution_builds_on_memory\": true\n  }\n}\n```\n\nField semantics:\n\n- `score` (1–10): how novel this run is relative to the accumulated\n  memory.\n- `compared_to_prior_runs`: 1–3 sentences comparing to specific prior\n  runs.\n- `contribution_builds_on_memory`: `true` if the research question\n  extends or responds to open questions or strong predictors from prior\n  runs.\n\nThis section is **optional**. Omit it entirely if no `## Findings Memory\nSummary` is in the user message. Do not include `\"novelty_review\": null`\n— simply omit the key.\n\n## Verification rules (Critic checklist `pf_08`, `pf_09`)\n\nWhen findings memory is enabled (the orchestrator sets a flag when the\nmemory file is non-empty):\n\n1. `pf_08` (minor): if the outcome variable and a substantially similar\n   predictor set have already been studied in a prior run (per\n   `studied_outcomes`), flag as minor unless the research question offers\n   a meaningfully different framing or predictor combination. Use\n   `novelty_review.score` as supporting evidence.\n2. `pf_09` (minor): verify that the research question or\n   `expected_contribution` references open questions or strong predictors\n   from the findings memory when relevant. Aspirational — minor, not\n   blocking.\n\n## Source provenance\n\nCanonical source: `agent_prompts\u002Fproblem_formulator.yaml` §\"Using\nFindings Memory\" + §\"Generating Diverse Candidates\".\n\nMerged content from:\n- `agent_prompts\u002Fcritic.yaml` §\"Optional: Novelty Review Against Prior\n  Runs\"\n- `data_registry\u002Fevaluation_rubrics\u002Fmethodological_checklist.yaml` items\n  `pf_08`, `pf_09`\n",{"name":5,"layer":24,"version":34,"priority":35,"resources":36,"description":7,"trigger_keywords":37,"applicable_stages":41,"references_skills":44,"applicable_datasets":45,"applicable_task_types":46},"1.0",3,[],[38,39,40,28],"findings-memory","prior-runs","novelty",[42,43],"ProblemFormulator","Critic",[],[],[],"main","1aeaead66d2bf876109f3b101005d2eed54041d9","skills\u002Fmethodology\u002Ffindings-memory-novelty-cross-run",0,"https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public","https:\u002F\u002Fgithub.com\u002Fcgpan","https:\u002F\u002Fcpan.ai","Ph.D. candidate in Measurement, Evaluation, and Statistics.","2026-09-28 20:39:06",{"source":8,"skillName":5,"commands":57,"prompt":67,"zipAvailable":68},[58,61,64],{"tool":59,"command":60},"npx","npx skills add cgpan\u002Fedm-ars-public --skill findings-memory-novelty-cross-run",{"tool":62,"command":63},"bunx","bunx skills add cgpan\u002Fedm-ars-public --skill findings-memory-novelty-cross-run",{"tool":65,"command":66},"pnpm","pnpm dlx skills add cgpan\u002Fedm-ars-public --skill findings-memory-novelty-cross-run","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public\u002Fblob\u002Fmain\u002Fskills\u002Fmethodology\u002Ffindings-memory-novelty-cross-run\u002FSKILL.md",true,[],[71,94,108,117,129,140],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":87,"publishedAt":29},"01M6EM1ZTQZVAP90DZTVBHCCFB","acm-fixture","Fixture writing skill for ACM acmart sigconf paper output.","https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Fwriting\u002Facm-fixture\u002FSKILL.md",[77,82],{"categoryNo":78,"categoryName":79,"taxonomyType":80,"taxonomyTypeDesc":81},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",{"categoryNo":83,"categoryName":84,"taxonomyType":85,"taxonomyTypeDesc":86},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[88,89,90,91,92,93],"acm fixture","fixture","writing","LaTeX","writer","paper",{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":29},"01M6EM1XNFETAZFGPHQPAXKBYN","prediction-fixture","Fixture skill for the prediction task type; references shap-fixture.","https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Ftask-type\u002Fprediction-fixture\u002FSKILL.md",[],[89,101,102,103,104,105,106,107],"prediction fixture","task","type","references","fixture skill","shap","analyst",{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":29},"01M6EM1V4MQXYGQ7K2ZDJRAPSM","causal-fixture","Fixture skill for the causal_inference task type.","https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Ftask-type\u002Fcausal-fixture\u002FSKILL.md",[],[89,115,103,102,116,105],"causal fixture","inference",{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":29},"01M6EM1SCWFZZ2ACSAP65GNWK9","shap-fixture","Fixture methodology for SHAP feature importance analysis.","https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Fmethodology\u002Fshap-fixture\u002FSKILL.md",[],[124,89,106,125,24,126,127,128],"shap fixture","feature importance","analysis","explainer","feature",{"skillNo":130,"name":131,"nameZh":6,"description":132,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":133,"categories":134,"keywords":135,"publishedAt":29},"01M6EM1PZC1MK1G0D1YFVYAQ9W","cluster-recon-fixture","Fixture methodology for reconstructing cluster IDs from fingerprint variables.","https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Fmethodology\u002Fcluster-recon-fixture\u002FSKILL.md",[],[136,137,138,89,139,24],"recon","cluster recon fixture","cluster","ids",{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":29},"01M6EM1MZVKMSWEXZ7TJ45DS04","hsls09-fixture","Fixture skill for HSLS:09 with a bundled YAML resource and one cross-reference.","https:\u002F\u002Fgithub.com\u002Fcgpan\u002Fedm-ars-public\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Fdataset\u002Fhsls09-fixture\u002FSKILL.md",[],[89,147,148,149,150,105,151,136,138,152],"hsls09 fixture","yaml","public","reference","dataset","registry"]