[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M74MP474ZA8F430V7PHZHKAX":3,"skill-same:01M74MP474ZA8F430V7PHZHKAX":53},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M74MP474ZA8F430V7PHZHKAX","discover-academic-skills",null,"Discover research\u002Facademic agent-skills from skills.sh (and user-supplied candidates), filter them through strict hard gates plus a strict academic-relevance gate, score survivors with a rubric, and hand back a ranked report with reasons. Use when the user wants to find, discover, or scout new academic\u002Fresearch skills to add to the hub, evaluate a skill they saw on social media, or refresh the discovery ledger.","pengqianhan\u002FAI-Human-Research-OS","pengqianhan","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F4286531?v=4","01M74M3DT9HNZ6HPAK8TX69EX9","PengqianHan","01M74M3DTVSN2GXW0JH2W396FK",3,"MIT","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fopen-paper-skills\u002Fdiscover-academic-skills\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: discover-academic-skills\ndescription: Discover research\u002Facademic agent-skills from skills.sh (and user-supplied candidates), filter them through strict hard gates plus a strict academic-relevance gate, score survivors with a rubric, and hand back a ranked report with reasons. Use when the user wants to find, discover, or scout new academic\u002Fresearch skills to add to the hub, evaluate a skill they saw on social media, or refresh the discovery ledger.\n---\n\n# Discover Academic Skills\n\nA **discovery scout**, not an auto-collector. It surfaces candidate research\nskills with scores and reasons; **ingestion is always human-gated** (per\n[ADR 0001](..\u002F..\u002Fdocs\u002Fadr\u002F0001-external-skill-intake-and-sync.md)). The scout\nnever installs or vendors anything.\n\n- **Scope**: discovery + strict filtering + scoring of academic\u002Fresearch skills\n  from the skills.sh registry and from candidates the user pastes (e.g. from\n  social media). Anything the user accepts is later vendored **by hand** into\n  [`collected-skills\u002F`](..\u002F..\u002Fcollected-skills\u002Findex.md) via the normal intake.\n- **Inputs**: none required (runs a default academic query set); optional\n  `--candidate owner\u002Frepo@skill`, `--queries`, `--min-installs`, `--include-seen`.\n- **Outputs**: a ranked Markdown report in `scratch\u002Fskill-discovery\u002F` (gitignored),\n  the top picks surfaced inline, and new rows appended to the committed\n  [`discovery-ledger.md`](..\u002F..\u002Fdiscovery-ledger.md).\n- **Limitations**: `npx skills find` is fuzzy and skills.sh coverage is partial;\n  the strict relevance gate favours precision over recall (it *will* miss some\n  real-but-unlabelled skills — by design, to cut review burden). The skills.sh\n  JSON API (`\u002Fapi\u002Fv1`) needs a Vercel OIDC token and is **not** used; the\n  unauthenticated CLI is the discovery channel.\n\n## Prerequisites\n\n- **Node \u002F npx** — for `npx skills find` (unauthenticated; no token).\n- **`gh` authenticated** — for the license hard gate and repo metadata. Without\n  it the license gate is skipped and every candidate is flagged\n  `license-unverified` for you to check by hand (nothing is silently dropped).\n\n## Pipeline (deterministic gates → relevance gate → rubric)\n\n### 1. Gather + deterministic hard gates (script)\n\nRun the helper — it queries skills.sh, dedupes, suppresses ledger entries, and\napplies the deterministic gates:\n\n```bash\npython research-skills-hub\u002Fopen-paper-skills\u002Fdiscover-academic-skills\u002Fscripts\u002Fdiscover.py\n```\n\nDeterministic gates it applies (a failure drops the candidate into the report's\n\"dropped\" appendix with a reason — never silently):\n\n- **Install floor** — `\u003C 50 installs` dropped (tune with `--min-installs`; manual\n  `--candidate` entries bypass this).\n- **License present** — no license → dropped (default all-rights-reserved can't be\n  redistributed). Permissive (MIT\u002FApache\u002FBSD\u002F…) passes; copyleft passes **flagged**\n  `copyleft-license`; unrecognised passes flagged for your check.\n- **Ledger suppression** — anything already decided in `discovery-ledger.md` is\n  dropped (use `--include-seen` to override). This is what makes repeated runs\n  quieter over time.\n- **Hub-overlap hint** — candidates whose name overlaps an existing hub skill are\n  flagged `possible-overlap:\u003Cname>` (not dropped — you may want the better one).\n\nThe script writes `scratch\u002Fskill-discovery\u002F\u003Cdate>-candidates.raw.json` with a\n`kept` list (for scoring) and a `dropped` list (for the appendix).\n\n### 2. Academic-relevance gate — STRICT (agent)\n\nFor every `kept` candidate, read its `SKILL.md` (open the skills.sh page at the\ncandidate `url`, or run `npx skills use owner\u002Frepo@skill` to print the skill\nprompt) and decide, strictly:\n\n> **Would a researcher use this *in a research workflow*?** It must clearly match\n> the IN list. Borderline or merely dual-use → **reject** (move to a\n> \"dropped — off scope\" section with a one-line reason).\n\n**IN (pass):** literature discovery\u002Freading (arXiv, OpenAlex, Semantic Scholar,\nPubMed, bioRxiv); citation\u002Freference management (Zotero, BibTeX); academic writing\n(LaTeX papers, venue templates, submission formatting); research data \u002F experiments\n\u002F reproducibility \u002F baseline tables; **paper-oriented** figures and diagrams;\nsystematic review \u002F meta-analysis \u002F bibliometrics; discipline-specific science\ntooling and ML\u002FAI research workflows.\n\n**OUT (reject):** general software engineering, web\u002Ffrontend, DevOps, marketing,\ngeneric productivity. **Dual-use tools** (generic `latex`, `diagram`, `pdf`,\n`search`) are rejected unless the SKILL.md gives an explicit research framing.\n\nPrecision over recall is intended: a strict gate keeps your review list almost\nentirely genuine research skills. The vocabulary behind this gate is in\n[`..\u002F..\u002FCONTEXT.md`](..\u002F..\u002FCONTEXT.md).\n\n### 3. Rubric scoring (agent)\n\nScore each survivor on four dimensions, 0–5 each, weighted into a 0–100 composite:\n\n| Dimension | Weight | What to judge |\n|---|---|---|\n| Research value & impact | 35% | how central\u002Fuseful to a research workflow (not \"is it relevant\" — \"how much does it matter\") |\n| Quality & maturity | 25% | SKILL.md clarity\u002Fcompleteness, install count, stars, recency of `pushed_at` |\n| Safety & trust | 25% | read the bundled scripts: does it run arbitrary code, reach the network, need credentials? author track record |\n| Fit & uniqueness | 15% | fills a gap vs the hub \u002F complements the OS; overlap with an existing skill lowers this |\n\nTiers by composite: **≥80 strongly recommend · 65–79 consider · 50–64 borderline\n· \u003C50 low (appendix only)**. License is not scored (it is a gate\u002Fflag).\n\n### 4. Report + ledger (agent)\n\nWrite `scratch\u002Fskill-discovery\u002F\u003Cdate>-candidates.md`:\n\n- Grouped by tier (strongly recommend → consider → borderline). Each entry:\n  composite score, the four sub-scores, a 2–3 sentence reason, and any flags\n  (`copyleft-license`, `license-\u003Cspdx>`, `possible-overlap:…`, `license-unverified`).\n- Appendices: **dropped at hard gates** and **dropped — off scope**, each with the\n  one-line reason. Do not silently omit anything.\n\nThen **surface every ≥80 candidate inline** to the user, and **append every\nsurfaced candidate** to [`discovery-ledger.md`](..\u002F..\u002Fdiscovery-ledger.md) with\ndisposition `surfaced`. When the user later accepts\u002Frejects one, update its row to\n`accepted` (note where it was vendored) or `rejected` (note why). Ledger row shape:\n\n```\n| owner\u002Frepo@skill | \u003Cdate> | surfaced | \u003Cscore> | \u003Cone-line reason> |\n```\n\n## Accepting a candidate\n\nAccepting is manual and follows the existing cherry-pick intake: read the whole\n`SKILL.md` and every bundled script, confirm the license, copy it into\n`collected-skills\u002F\u003Cskill>\u002F`, register it with the `skill-organizer` skill\n(source URL + license), then run `.\u002Fverify.sh`. If a whole repo is worth tracking\nwholesale, promote it to its own vendored-mirror collection like `mattpocock-skills`\ninstead (see ADR 0001).\n\n## Manual candidate (social-media finds)\n\nFeed a skill you saw elsewhere through the exact same gates and rubric:\n\n```bash\npython ...\u002Fdiscover.py --candidate owner\u002Frepo@skill\n```\n\nIt skips the install floor but runs the license gate, relevance gate, and rubric\nlike any other candidate.\n","# Discover Academic Skills\n\nA **discovery scout**, not an auto-collector. It surfaces candidate research\nskills with scores and reasons; **ingestion is always human-gated** (per\n[ADR 0001](..\u002F..\u002Fdocs\u002Fadr\u002F0001-external-skill-intake-and-sync.md)). The scout\nnever installs or vendors anything.\n\n- **Scope**: discovery + strict filtering + scoring of academic\u002Fresearch skills\n  from the skills.sh registry and from candidates the user pastes (e.g. from\n  social media). Anything the user accepts is later vendored **by hand** into\n  [`collected-skills\u002F`](..\u002F..\u002Fcollected-skills\u002Findex.md) via the normal intake.\n- **Inputs**: none required (runs a default academic query set); optional\n  `--candidate owner\u002Frepo@skill`, `--queries`, `--min-installs`, `--include-seen`.\n- **Outputs**: a ranked Markdown report in `scratch\u002Fskill-discovery\u002F` (gitignored),\n  the top picks surfaced inline, and new rows appended to the committed\n  [`discovery-ledger.md`](..\u002F..\u002Fdiscovery-ledger.md).\n- **Limitations**: `npx skills find` is fuzzy and skills.sh coverage is partial;\n  the strict relevance gate favours precision over recall (it *will* miss some\n  real-but-unlabelled skills — by design, to cut review burden). The skills.sh\n  JSON API (`\u002Fapi\u002Fv1`) needs a Vercel OIDC token and is **not** used; the\n  unauthenticated CLI is the discovery channel.\n\n## Prerequisites\n\n- **Node \u002F npx** — for `npx skills find` (unauthenticated; no token).\n- **`gh` authenticated** — for the license hard gate and repo metadata. Without\n  it the license gate is skipped and every candidate is flagged\n  `license-unverified` for you to check by hand (nothing is silently dropped).\n\n## Pipeline (deterministic gates → relevance gate → rubric)\n\n### 1. Gather + deterministic hard gates (script)\n\nRun the helper — it queries skills.sh, dedupes, suppresses ledger entries, and\napplies the deterministic gates:\n\n```bash\npython research-skills-hub\u002Fopen-paper-skills\u002Fdiscover-academic-skills\u002Fscripts\u002Fdiscover.py\n```\n\nDeterministic gates it applies (a failure drops the candidate into the report's\n\"dropped\" appendix with a reason — never silently):\n\n- **Install floor** — `\u003C 50 installs` dropped (tune with `--min-installs`; manual\n  `--candidate` entries bypass this).\n- **License present** — no license → dropped (default all-rights-reserved can't be\n  redistributed). Permissive (MIT\u002FApache\u002FBSD\u002F…) passes; copyleft passes **flagged**\n  `copyleft-license`; unrecognised passes flagged for your check.\n- **Ledger suppression** — anything already decided in `discovery-ledger.md` is\n  dropped (use `--include-seen` to override). This is what makes repeated runs\n  quieter over time.\n- **Hub-overlap hint** — candidates whose name overlaps an existing hub skill are\n  flagged `possible-overlap:\u003Cname>` (not dropped — you may want the better one).\n\nThe script writes `scratch\u002Fskill-discovery\u002F\u003Cdate>-candidates.raw.json` with a\n`kept` list (for scoring) and a `dropped` list (for the appendix).\n\n### 2. Academic-relevance gate — STRICT (agent)\n\nFor every `kept` candidate, read its `SKILL.md` (open the skills.sh page at the\ncandidate `url`, or run `npx skills use owner\u002Frepo@skill` to print the skill\nprompt) and decide, strictly:\n\n> **Would a researcher use this *in a research workflow*?** It must clearly match\n> the IN list. Borderline or merely dual-use → **reject** (move to a\n> \"dropped — off scope\" section with a one-line reason).\n\n**IN (pass):** literature discovery\u002Freading (arXiv, OpenAlex, Semantic Scholar,\nPubMed, bioRxiv); citation\u002Freference management (Zotero, BibTeX); academic writing\n(LaTeX papers, venue templates, submission formatting); research data \u002F experiments\n\u002F reproducibility \u002F baseline tables; **paper-oriented** figures and diagrams;\nsystematic review \u002F meta-analysis \u002F bibliometrics; discipline-specific science\ntooling and ML\u002FAI research workflows.\n\n**OUT (reject):** general software engineering, web\u002Ffrontend, DevOps, marketing,\ngeneric productivity. **Dual-use tools** (generic `latex`, `diagram`, `pdf`,\n`search`) are rejected unless the SKILL.md gives an explicit research framing.\n\nPrecision over recall is intended: a strict gate keeps your review list almost\nentirely genuine research skills. The vocabulary behind this gate is in\n[`..\u002F..\u002FCONTEXT.md`](..\u002F..\u002FCONTEXT.md).\n\n### 3. Rubric scoring (agent)\n\nScore each survivor on four dimensions, 0–5 each, weighted into a 0–100 composite:\n\n| Dimension | Weight | What to judge |\n|---|---|---|\n| Research value & impact | 35% | how central\u002Fuseful to a research workflow (not \"is it relevant\" — \"how much does it matter\") |\n| Quality & maturity | 25% | SKILL.md clarity\u002Fcompleteness, install count, stars, recency of `pushed_at` |\n| Safety & trust | 25% | read the bundled scripts: does it run arbitrary code, reach the network, need credentials? author track record |\n| Fit & uniqueness | 15% | fills a gap vs the hub \u002F complements the OS; overlap with an existing skill lowers this |\n\nTiers by composite: **≥80 strongly recommend · 65–79 consider · 50–64 borderline\n· \u003C50 low (appendix only)**. License is not scored (it is a gate\u002Fflag).\n\n### 4. Report + ledger (agent)\n\nWrite `scratch\u002Fskill-discovery\u002F\u003Cdate>-candidates.md`:\n\n- Grouped by tier (strongly recommend → consider → borderline). Each entry:\n  composite score, the four sub-scores, a 2–3 sentence reason, and any flags\n  (`copyleft-license`, `license-\u003Cspdx>`, `possible-overlap:…`, `license-unverified`).\n- Appendices: **dropped at hard gates** and **dropped — off scope**, each with the\n  one-line reason. Do not silently omit anything.\n\nThen **surface every ≥80 candidate inline** to the user, and **append every\nsurfaced candidate** to [`discovery-ledger.md`](..\u002F..\u002Fdiscovery-ledger.md) with\ndisposition `surfaced`. When the user later accepts\u002Frejects one, update its row to\n`accepted` (note where it was vendored) or `rejected` (note why). Ledger row shape:\n\n```\n| owner\u002Frepo@skill | \u003Cdate> | surfaced | \u003Cscore> | \u003Cone-line reason> |\n```\n\n## Accepting a candidate\n\nAccepting is manual and follows the existing cherry-pick intake: read the whole\n`SKILL.md` and every bundled script, confirm the license, copy it into\n`collected-skills\u002F\u003Cskill>\u002F`, register it with the `skill-organizer` skill\n(source URL + license), then run `.\u002Fverify.sh`. If a whole repo is worth tracking\nwholesale, promote it to its own vendored-mirror collection like `mattpocock-skills`\ninstead (see ADR 0001).\n\n## Manual candidate (social-media finds)\n\nFeed a skill you saw elsewhere through the exact same gates and rubric:\n\n```bash\npython ...\u002Fdiscover.py --candidate owner\u002Frepo@skill\n```\n\nIt skips the install floor but runs the license gate, relevance gate, and rubric\nlike any other candidate.\n",{"name":5,"description":7},"main","e71245365df78e06cd166602744bf2491f79910e","research-skills-hub\u002Fopen-paper-skills\u002Fdiscover-academic-skills",0,"https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS","https:\u002F\u002Fpengqianhan.github.io\u002FAI-Human-Research-OS\u002F","https:\u002F\u002Fgithub.com\u002Fpengqianhan","Ph.D. candidate of The University of Auckland, New Zealand","2026-11-14 18:08:16",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add pengqianhan\u002FAI-Human-Research-OS --skill discover-academic-skills",{"tool":39,"command":40},"bunx","bunx skills add pengqianhan\u002FAI-Human-Research-OS --skill discover-academic-skills",{"tool":42,"command":43},"pnpm","pnpm dlx skills add pengqianhan\u002FAI-Human-Research-OS --skill discover-academic-skills","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fopen-paper-skills\u002Fdiscover-academic-skills\u002FSKILL.md",true,[47],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"scripts\u002Fdiscover.py","SCRIPT","脚本",12137,"https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fopen-paper-skills\u002Fdiscover-academic-skills\u002Fscripts\u002Fdiscover.py",[54,61,68,75,82,89],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M74MWYTHT498SVMN823896T7","workflow-skill-creator","Distills a completed user workflow or interaction into a reusable agent skill. Use when the user asks to turn their workflow, interaction, or multi-step process into a skill, or when they say \"make this a skill\", \"create a skill from what we just did\", \"package this workflow\" or similar. Do not use for creating skills from scratch without an existing workflow (use a generic skill-creator for that).","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Fworkflow_skill_creator\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M74MWWVCFZ8HQS8840F1V1FT","uv","Checks whether the uv Python package manager is installed and installs it if missing. Ensures uv is on PATH. Use when another skill requires uv as a prerequisite.","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Fuv\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M74MWRH5BTSWZ2C5PCTD5FV9","uniprot-database","Access protein metadata, function, taxonomy, and sequences across UniProtKB, UniParc, and UniRef. Use when searching for proteins, mapping identifiers, or retrieving functional annotations and publications. Don't use for sequence alignment, protein folding, or sequence similarity search (use specialized skills for those tasks).","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Funiprot_database\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M74MWN1D3ER9F1FZPWH8HQC1","unibind-database","Queries the UniBind database for experimentally validated transcription factor (TF) binding sites. Use when retrieving direct TF-DNA interaction datasets, downloading binding site coordinates (BED\u002FFASTA) for local analysis, or listing available datasets by species, cell line, or TF name. Don't use to query specific intervals, locations, genes, motif models or expression data.","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Funibind_database\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M74MWHHB5QZY3NR2FYA1EK4N","ucsc-conservation-and-tfbs","Fetch Evolutionary Conservation scores (phyloP, phastCons) and Transcription Factor Binding Sites (TFBS) from the UCSC Genome Browser. Use when analyzing whether genomic variants or regions are evolutionarily conserved, functionally important, or bounded by TF regulators across major projects (ENCODE, JASPAR, ReMap).","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Fucsc_conservation_and_tfbs\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M74MWERYHSY87P21FDMVE6BA","string-database","Query the STRING database for protein-protein interactions (PPIs), functional enrichment, and homology. Use when the user asks about interactions between specific proteins, interaction evidence, confidence scores, protein interaction partners, or pathway enrichments.","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Fstring_database\u002FSKILL.md",[],[]]