[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DCN53EK6RGT2JTEMADNF7C":3,"skill-same:01M7DCN53EK6RGT2JTEMADNF7C":60},{"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":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M7DCN53EK6RGT2JTEMADNF7C","clinvar-database",null,"Use when needing clinical significance, pathogenicity classifications (e.g., Pathogenic, Benign, VUS), clinical evidence rationales, or finding \"hard positive\" benchmark controls for human genomic variants.","Marvellbrazil\u002Fmy-agentic-skills","Marvellbrazil","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F184349613?v=4","01M7DCJPCRHP7SZS0T5H1GJZ8F","Marvello Faisal","01M7DCJPDCZDB94VG0QXGC6RC3",5,"MIT","https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Farchive\u002Fbioinformatics\u002Fclinvar_database\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: clinvar-database\ndescription: >\n  Use when needing clinical significance, pathogenicity classifications (e.g.,\n  Pathogenic, Benign, VUS), clinical evidence rationales, or finding \"hard\n  positive\" benchmark controls for human genomic variants.\n---\n\n# ClinVar Database\n\n## Prerequisites\n\n1.  **`uv`**: Read the `uv` skill and follow its Setup instructions to ensure\n    `uv` is installed and on PATH.\n2.  **User Notification**: If .licenses\u002Fclinvar_database_LICENSE.txt does not\n    already exist in the workspace root directory then (1) prominently notify\n    the user to check the terms at https:\u002F\u002Fwww.ncbi.nlm.nih.gov\u002Fclinvar\u002F, then\n    (2) create the file recording the notification text and timestamp.\n3.  **`.env` file**: Make sure the `.env` file exists in your home directory.\n    Create one if it does not exist.\n4.  **`NCBI_API_KEY`** (optional): Raises the NCBI rate limit from 3 to 10\n    requests\u002Fsecond. The skill works without it, but a key is recommended if the\n    user plans many queries or encounters a 429 error. You can register for a\n    key for free at https:\u002F\u002Fwww.ncbi.nlm.nih.gov\u002Faccount\u002Fsettings\u002F. You **MUST**\n    use the safe credentials protocol in the `credentials` skill to check for\n    and request this key if this skill looks relevant to the user's request.\n\n## Overview\n\nClinVar is the primary consensus record for clinical classifications of human\ngenomic variations. It provides the \"clinical ground truth\" for pathogenicity\nlabels (Pathogenic, Likely Pathogenic, Benign, VUS) based on assertions from\nglobal laboratories.\n\n## When to Use\n\n**Use when you need to:**\n\n-   Find the current clinical significance and star rating (review status) for a\n    specific variant.\n-   Fetch clinician notes, assertion criteria, or rationales for previous\n    clinical laboratory classifications.\n-   Retrieve the preferred condition name and associated HPO terms for a\n    specific variant.\n-   Find a list of variant controls (e.g., \"Find all Pathogenic variants in the\n    HBB gene within 50bp of a signal\").\n-   Check for conflicting interpretations for a given variant and identify the\n    organizations submitting each classification.\n\n**Do NOT use when you need to:**\n\n-   Find specific allele frequencies in global populations (use **gnomAD**).\n-   Describe the normal biological role of a protein and typical inheritance\n    patterns (use **OMIM**).\n-   Predict mechanistic effects of novel mutations, like frameshifts or exon\n    skipping (use **AlphaGenome**).\n-   Find recommended surveillance schedules for patients with a pathogenic\n    variant (use **GeneReviews**).\n-   Generate or view 3D structural models of affected proteins (use **PDB \u002F\n    AlphaFold**).\n\n## Quick Start\n\nClinVar queries are executed via a robust Python wrapper script to handle strict\nrate limiting and XML\u002FJSON parsing.\n\nExample: Search for BRCA1 variants\n\n```bash\nuv run scripts\u002Fclinvar_api.py search --query \"BRCA1[gene]\" --output results.json\n```\n\n## Core Rules\n\n-   **Retmax Constraint**: The search command defaults to `--retmax 200`. For\n    any \"List all\" or gene-wide request, you MUST explicitly set `--retmax`\n    higher (e.g., 1000) to ensure data completeness.\n-   **Use the Wrapper**: Prefer the wrapper script for standard queries. It\n    handles rate limiting, retries, and the complex XML parsing for you. If the\n    script's parsed output does not contain the specific fields you need, you\n    may modify the script or query the NCBI E-utilities API directly — but be\n    aware that the raw XML schemas are complex and vary between record types.\n-   If the rate limit is hit, the script will throw a clear error. You **MUST**\n    use the safe credentials protocol in the `credentials` skill to check for\n    and request the `NCBI_API_KEY` to help the user add it to their `.env` file.\n-   **Notification**: If this skill is used, ensure this is mentioned in the\n    output.\n\n## Utility Scripts\n\n### 1. `count` — Count Matching Variants\n\n**Purpose:** Check how many variants match a query without fetching IDs. Use to\ndecide whether a full `search` is warranted.\n\n*Arguments:*\n\n-   `--query`: (Required) NCBI Entrez search query string.\n-   `--output`: (Required) Output JSON file path.\n\n*Example:* `uv run scripts\u002Fclinvar_api.py count \\ --query \"TP53[gene] AND\n\\\"uncertain significance\\\"[clinsig]\" \\ --output count.json` *Output:*\n`{\"total_count\": \u003Cint>}`\n\n### 2. `search` — Search Variants\n\n**Purpose:** Identify variants based on genomic location, gene symbols, or\nclinical attributes using NCBI Entrez search syntax. The search command\n**automatically paginates** through all matching results to ensure complete,\ndeterministic retrieval.\n\n```bash\n# Fetch ALL matching variants (default behavior)\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"BRCA1[gene]\" --output results.json\n\n# Search by Chromosome and Position Range\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"11[chr] AND 5225000:5226000[chrpos]\" --output results.json\n\n# Combine terms using Entrez syntax\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"HBB[gene] AND pathogenic[clinsig]\" --output results.json\n\n# Cap results at 50\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"TP53[gene]\" --retmax 50 --output results.json\n```\n\n*Arguments:*\n\n-   `--query`: (Required) NCBI Entrez search query string.\n-   `--retmax`: Maximum total number of variant IDs to return. **Default is 0,\n    which means \"fetch all matching results.\"** Set to a positive integer to cap\n    the result set.\n-   `--page_size`: Number of IDs to fetch per API request (default: 500, max:\n    10000 per NCBI limits).\n-   `--output`: (Required) Output JSON file path.\n\n*Output:* A JSON object containing:\n\n-   `total_count` — Total number of matching variants in ClinVar.\n-   `fetched_count` — Number of IDs actually retrieved.\n-   `variant_ids` — List of ClinVar Variation ID strings.\n\n### 3. `summary` — Get Interpretation Summary\n\n**Purpose:** Retrieve top-line clinical significance labels, star ratings\n(review status), and basic phenotype data for rapid variant screening.\n\n```bash\n# Get summary for one or more Variation IDs\nuv run scripts\u002Fclinvar_api.py summary \\\n  --variant_ids 12345 67890 --output summary.json\n```\n\n*Arguments:*\n\n-   `--variant_ids`: (Required) One or more ClinVar Variation IDs.\n-   `--output`: (Required) Output JSON file path.\n\n*Output:* A JSON list of summary objects, each containing:\n\n-   `variant_id`, `title`, `clinical_significance`, `review_status`, \\\n    `last_evaluated`, `phenotypes`\n-   `genes` — list of `{gene_id, symbol, strand}`\n-   `variation_type` — e.g., single nucleotide variant, Deletion, Insertion\n-   `molecular_consequences` — list of strings (e.g., [\"missense variant\", \\\n    \"nonsense\"])\n\n### 4. `evidence` — Get Clinical Evidence\n\n**Purpose:** Fetch the full clinical record for a single variant, including\nfree-text clinician rationales, assertion methods, and specific submitter notes.\n\n```bash\n# Get full evidence for a single Variation ID\nuv run scripts\u002Fclinvar_api.py evidence \\\n  --variant_id 12345 --output evidence.json\n```\n\n*Arguments:*\n\n-   `--variant_id`: (Required) A single ClinVar Variation ID.\n-   `--output`: (Required) Output JSON file path.\n\n*Output:* A JSON object containing:\n\n-   `variant_id`\n-   `allele_info` — `{chromosome, position_start, position_stop,\n    reference_allele, alternate_allele, cytogenetic_band, dbsnp_rsid}` (GRCh38\n    preferred)\n-   `conditions` — list of `{name, medgen_cui, omim_id, orphanet_id, hpo_terms}`\n-   `functional_consequences` — list of `{value, sequence_ontology_id}`\n-   `structural_variant_details` — `{outer_start, inner_start, inner_stop,\n    outer_stop, copy_number}` (present only for CNVs, otherwise null)\n-   `citation_references` — list of PubMed IDs cited in the global \"Citations\"\n    section\n-   `submissions` — list of per-submitter records, each containing:\n    -   `submitter_name`, `classification`, `curator_notes`,\n        `assertion_criteria`\n    -   `date_last_evaluated` — when the submitter last reviewed the\n        classification\n\n## Typical Workflows\n\n### Count-First Workflow (Recommended)\n\nFor large or unknown result sets, use `count` first to decide whether to\nproceed, then `search` (which auto-paginates and returns `total_count` \u002F\n`fetched_count`), then `summary` to screen.\n\n```bash\n# Step 1: Gauge size (optional — search also returns total_count)\nuv run scripts\u002Fclinvar_api.py count \\\n  --query \"HBB[gene] AND pathogenic[clinsig]\" --output count.json\n\n# Step 2: Fetch all variant IDs (auto-paginates)\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"HBB[gene] AND pathogenic[clinsig]\" --output ids.json\n\n# Step 3: Get summaries (extract variant_ids from search output)\nuv run scripts\u002Fclinvar_api.py summary \\\n  --variant_ids 12345 67890 --output summary.json\n```\n\n### Deep Dive: search → evidence\n\nWhen you need the full clinical picture for a specific variant — including\nsubmitter rationales, PubMed citations, ontology-linked conditions, and allele\ncoordinates — use `evidence`.\n\n```bash\nuv run scripts\u002Fclinvar_api.py evidence \\\n  --variant_id 12345 --output evidence.json\n```\n\n### Workflow: Robust Variant Discovery (Triangulation)\n\nClinVar metadata is inconsistent. To fulfill \"List all\" requests, do not rely on\na single filter. Perform the following in a single turn and merge results:\n\n1.  **Search by exact label** (e.g., `\"3 prime UTR\n    variant\"[molecular_consequence]`).\n2.  **Search by HGVS nomenclature pattern** (e.g., `c.*`).\n3.  **Search by genomic coordinate range** (using `[chrpos]`).\n\nThis \"triangulation\" ensures structural variants with missing labels are not\noverlooked.\n\n### Verifying Coding vs. Non-Coding Status via HGVS\n\n`molecular_consequences` alone can be ambiguous (e.g., `splice donor variant`\nappears in both coding and non-coding contexts). Always cross-check the `title`\nfield for HGVS patterns:\n\n-   `c.-…` — 5' UTR (non-coding)\n-   `c.*…` — 3' UTR (non-coding)\n-   `c.123+N` \u002F `c.123-N` — intronic (non-coding)\n-   `p.Trp146Arg` etc. — protein effect (coding)\n\nA variant with UTR\u002Fintronic HGVS and no `p.` annotation is non-coding, even with\nsplicing labels. Conversely, any `p.` annotation indicates a coding effect.\n\n### ClinVar Metadata Reference\n\n-   **3' UTR**\n    -   Search String: `\"3 prime UTR variant\"[mol_consequence]`\n    -   HGVS: `c.*`\n-   **5' UTR**\n    -   Search String: `\"5 prime UTR variant\"[mol_consequence]`\n    -   HGVS: `c.-`\n-   To find \"high-confidence\" variants or expert-reviewed consensus, use the\n    `review_status` filter. This is the most efficient way to distinguish\n    between single-laboratory assertions and panel-reviewed ground truth.\n\n### When to Use Which Fields\n\n-   **Quick pathogenicity label** — Use `summary` → `clinical_significance`\n-   **Gene symbol and strand** — Use `summary` → `genes`\n-   **Variant type (SNV, del, etc.)** — Use `summary` → `variation_type`\n-   **Protein-level effect** — Use `summary` → `molecular_consequences`\n-   **Genomic coordinates (GRCh38)** — Use `evidence` → `allele_info`\n-   **Linked conditions (ontology)** — Use `evidence` → `conditions`\n-   **SO functional consequence** — Use `evidence` → `functional_consequences`\n-   **CNV breakpoints\u002Fcopy number** — Use `evidence` →\n    `structural_variant_details`\n-   **PubMed references** — Use `evidence` → `citation_references`\n-   **Date of last lab review** — Use both → `last_evaluated`\n-   **Clinician rationales** — Use `evidence` → `submissions[].curator_notes`\n\n### Retrieving Genomic Coordinates (Default HG38\u002FGRCh38)\n\nTo get precise genomic coordinates in the format `\u003Cchrom>:\u003Cpos>:\u003Cref>>\u003Calt>`\n(e.g., `chr5:70951945:G>A`), you must use the `evidence` command, as these\ndetails are not available in the `summary` output.\n\n**You MUST always include genomic coordinates in the format\n`\u003Cchrom>:\u003Cpos>:\u003Cref>>\u003Calt>` when listing or presenting variants, even if not\nexplicitly requested by the user. If coordinates are missing from the summary,\nuse the `evidence` command or dbSNP fallback to retrieve them.**\n\n1.  **Fetch Evidence**: Use `uv run scripts\u002Fclinvar_api.py evidence --variant_id\n    \u003CID> --output evidence.json`.\n2.  **Extract VCF Attributes**: The `evidence` command parses the XML. Extract:\n    *   Chromosome: `Chr`\n    *   Position: `positionVCF` (or `start`)\n    *   Ref: `referenceAlleleVCF` (or `referenceAllele`)\n    *   Alt: `alternateAlleleVCF` (or `alternateAllele`) from the\n        `SequenceLocation` element with `Assembly=\"GRCh38\"`.\n\n**Fallback for Imprecise Coordinates (Gene Range):** ClinVar often returns the\nfull gene range for non-coding variants. If the extracted coordinates correspond\nto the gene range instead of a specific position, use the `dbsnp-database` skill\nto resolve the precise coordinates using the `dbsnp_rsid` or HGVS title: 1.Check\nfor `dbsnp_rsid` in the `evidence` output. 2. Run `uv run scripts\u002Fdbsnp_cli.py\nresolve-rsid {rsid}` to get precise GRCh38 coordinates. 3. Format as\n`\u003Cchrom>:\u003Cpos>:\u003Cref>>\u003Calt>` using the SPDI or HGVS data from dbSNP.\n\n### Structural Variant Note\n\nThe `structural_variant_details` field is **only populated for copy number\nvariants (CNVs)**. For standard SNVs and small indels this field will be `null`.\nUse the `allele_info` fields (`position_start`, `position_stop`,\n`reference_allele`, `alternate_allele`) instead.\n\n### CNV \u002F Large Deletion Note\n\nLarge copy-number variants (CNVs) frequently have empty\n`molecular_consequences`. If a variant title mentions \"del\" and coordinates\noverlap your target region, it is relevant regardless of missing labels.\n\n### Obtaining and Using an API Key\n\nYou can register for a key for free at\nhttps:\u002F\u002Fwww.ncbi.nlm.nih.gov\u002Faccount\u002Fsettings\u002F. You **MUST** use the safe\ncredentials protocol in the `credentials` skill to check for and request this\nkey if this skill looks relevant to the user's request.\n\n## Best Practices\n\n-   Always use `uv run` to execute `python`.\n-   If `jq` is unavailable pivot immediately to using Python one-liners for\n    processing JSON (e.g., `uv run python3 -c \"import json; ...\"`).\n-   Use `count` before `search` to understand the result set size.\n-   The `search` command fetches all results by default and includes\n    `total_count` and `fetched_count` in the output — always verify these match\n    to confirm complete retrieval.\n-   Entrez results are **unsorted**. To order by date, fetch all results and\n    sort locally by `last_evaluated`.\n\n## Common Mistakes\n\n-   **Attempting to parse the E-utilities XML yourself** — Always use the\n    provided `clinvar_api.py` client which handles the unpredictable XML schemas\n    robustly.\n-   **Getting HTTP 429 Too Many Requests** — The client throws an exception\n    telling you to pause. You **MUST** use the safe credentials protocol in the\n    `credentials` skill to check for and request the `NCBI_API_KEY` to help the\n    user add it to their `.env` file, then retry.\n-   **Sending raw DNA sequences to the API** — The API expects HGVS\n    nomenclature, RS IDs, or proper Entrez coordinate syntax (`11[chr] AND\n    1234[chrpos]`), not raw ATCG strings.\n-   **For synonymous or non-coding variants** — HGVS nomenclature (e.g., CAPN3\n    AND \"c.551C>T\") is more reliable than coordinate searches ([chrpos]), as\n    many ClinVar records for these types lack precise genomic mappings.\n-   **Case sensitivity in molecular consequences** — ClinVar returns mixed-case\n    strings. Always use case-insensitive matching (`.lower()`) when filtering.\n-   **Parsing `search` output as a bare list** — `search` returns a JSON object\n    with `total_count`, `fetched_count`, and `variant_ids` — not a bare list.\n","# ClinVar Database\n\n## Prerequisites\n\n1.  **`uv`**: Read the `uv` skill and follow its Setup instructions to ensure\n    `uv` is installed and on PATH.\n2.  **User Notification**: If .licenses\u002Fclinvar_database_LICENSE.txt does not\n    already exist in the workspace root directory then (1) prominently notify\n    the user to check the terms at https:\u002F\u002Fwww.ncbi.nlm.nih.gov\u002Fclinvar\u002F, then\n    (2) create the file recording the notification text and timestamp.\n3.  **`.env` file**: Make sure the `.env` file exists in your home directory.\n    Create one if it does not exist.\n4.  **`NCBI_API_KEY`** (optional): Raises the NCBI rate limit from 3 to 10\n    requests\u002Fsecond. The skill works without it, but a key is recommended if the\n    user plans many queries or encounters a 429 error. You can register for a\n    key for free at https:\u002F\u002Fwww.ncbi.nlm.nih.gov\u002Faccount\u002Fsettings\u002F. You **MUST**\n    use the safe credentials protocol in the `credentials` skill to check for\n    and request this key if this skill looks relevant to the user's request.\n\n## Overview\n\nClinVar is the primary consensus record for clinical classifications of human\ngenomic variations. It provides the \"clinical ground truth\" for pathogenicity\nlabels (Pathogenic, Likely Pathogenic, Benign, VUS) based on assertions from\nglobal laboratories.\n\n## When to Use\n\n**Use when you need to:**\n\n-   Find the current clinical significance and star rating (review status) for a\n    specific variant.\n-   Fetch clinician notes, assertion criteria, or rationales for previous\n    clinical laboratory classifications.\n-   Retrieve the preferred condition name and associated HPO terms for a\n    specific variant.\n-   Find a list of variant controls (e.g., \"Find all Pathogenic variants in the\n    HBB gene within 50bp of a signal\").\n-   Check for conflicting interpretations for a given variant and identify the\n    organizations submitting each classification.\n\n**Do NOT use when you need to:**\n\n-   Find specific allele frequencies in global populations (use **gnomAD**).\n-   Describe the normal biological role of a protein and typical inheritance\n    patterns (use **OMIM**).\n-   Predict mechanistic effects of novel mutations, like frameshifts or exon\n    skipping (use **AlphaGenome**).\n-   Find recommended surveillance schedules for patients with a pathogenic\n    variant (use **GeneReviews**).\n-   Generate or view 3D structural models of affected proteins (use **PDB \u002F\n    AlphaFold**).\n\n## Quick Start\n\nClinVar queries are executed via a robust Python wrapper script to handle strict\nrate limiting and XML\u002FJSON parsing.\n\nExample: Search for BRCA1 variants\n\n```bash\nuv run scripts\u002Fclinvar_api.py search --query \"BRCA1[gene]\" --output results.json\n```\n\n## Core Rules\n\n-   **Retmax Constraint**: The search command defaults to `--retmax 200`. For\n    any \"List all\" or gene-wide request, you MUST explicitly set `--retmax`\n    higher (e.g., 1000) to ensure data completeness.\n-   **Use the Wrapper**: Prefer the wrapper script for standard queries. It\n    handles rate limiting, retries, and the complex XML parsing for you. If the\n    script's parsed output does not contain the specific fields you need, you\n    may modify the script or query the NCBI E-utilities API directly — but be\n    aware that the raw XML schemas are complex and vary between record types.\n-   If the rate limit is hit, the script will throw a clear error. You **MUST**\n    use the safe credentials protocol in the `credentials` skill to check for\n    and request the `NCBI_API_KEY` to help the user add it to their `.env` file.\n-   **Notification**: If this skill is used, ensure this is mentioned in the\n    output.\n\n## Utility Scripts\n\n### 1. `count` — Count Matching Variants\n\n**Purpose:** Check how many variants match a query without fetching IDs. Use to\ndecide whether a full `search` is warranted.\n\n*Arguments:*\n\n-   `--query`: (Required) NCBI Entrez search query string.\n-   `--output`: (Required) Output JSON file path.\n\n*Example:* `uv run scripts\u002Fclinvar_api.py count \\ --query \"TP53[gene] AND\n\\\"uncertain significance\\\"[clinsig]\" \\ --output count.json` *Output:*\n`{\"total_count\": \u003Cint>}`\n\n### 2. `search` — Search Variants\n\n**Purpose:** Identify variants based on genomic location, gene symbols, or\nclinical attributes using NCBI Entrez search syntax. The search command\n**automatically paginates** through all matching results to ensure complete,\ndeterministic retrieval.\n\n```bash\n# Fetch ALL matching variants (default behavior)\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"BRCA1[gene]\" --output results.json\n\n# Search by Chromosome and Position Range\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"11[chr] AND 5225000:5226000[chrpos]\" --output results.json\n\n# Combine terms using Entrez syntax\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"HBB[gene] AND pathogenic[clinsig]\" --output results.json\n\n# Cap results at 50\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"TP53[gene]\" --retmax 50 --output results.json\n```\n\n*Arguments:*\n\n-   `--query`: (Required) NCBI Entrez search query string.\n-   `--retmax`: Maximum total number of variant IDs to return. **Default is 0,\n    which means \"fetch all matching results.\"** Set to a positive integer to cap\n    the result set.\n-   `--page_size`: Number of IDs to fetch per API request (default: 500, max:\n    10000 per NCBI limits).\n-   `--output`: (Required) Output JSON file path.\n\n*Output:* A JSON object containing:\n\n-   `total_count` — Total number of matching variants in ClinVar.\n-   `fetched_count` — Number of IDs actually retrieved.\n-   `variant_ids` — List of ClinVar Variation ID strings.\n\n### 3. `summary` — Get Interpretation Summary\n\n**Purpose:** Retrieve top-line clinical significance labels, star ratings\n(review status), and basic phenotype data for rapid variant screening.\n\n```bash\n# Get summary for one or more Variation IDs\nuv run scripts\u002Fclinvar_api.py summary \\\n  --variant_ids 12345 67890 --output summary.json\n```\n\n*Arguments:*\n\n-   `--variant_ids`: (Required) One or more ClinVar Variation IDs.\n-   `--output`: (Required) Output JSON file path.\n\n*Output:* A JSON list of summary objects, each containing:\n\n-   `variant_id`, `title`, `clinical_significance`, `review_status`, \\\n    `last_evaluated`, `phenotypes`\n-   `genes` — list of `{gene_id, symbol, strand}`\n-   `variation_type` — e.g., single nucleotide variant, Deletion, Insertion\n-   `molecular_consequences` — list of strings (e.g., [\"missense variant\", \\\n    \"nonsense\"])\n\n### 4. `evidence` — Get Clinical Evidence\n\n**Purpose:** Fetch the full clinical record for a single variant, including\nfree-text clinician rationales, assertion methods, and specific submitter notes.\n\n```bash\n# Get full evidence for a single Variation ID\nuv run scripts\u002Fclinvar_api.py evidence \\\n  --variant_id 12345 --output evidence.json\n```\n\n*Arguments:*\n\n-   `--variant_id`: (Required) A single ClinVar Variation ID.\n-   `--output`: (Required) Output JSON file path.\n\n*Output:* A JSON object containing:\n\n-   `variant_id`\n-   `allele_info` — `{chromosome, position_start, position_stop,\n    reference_allele, alternate_allele, cytogenetic_band, dbsnp_rsid}` (GRCh38\n    preferred)\n-   `conditions` — list of `{name, medgen_cui, omim_id, orphanet_id, hpo_terms}`\n-   `functional_consequences` — list of `{value, sequence_ontology_id}`\n-   `structural_variant_details` — `{outer_start, inner_start, inner_stop,\n    outer_stop, copy_number}` (present only for CNVs, otherwise null)\n-   `citation_references` — list of PubMed IDs cited in the global \"Citations\"\n    section\n-   `submissions` — list of per-submitter records, each containing:\n    -   `submitter_name`, `classification`, `curator_notes`,\n        `assertion_criteria`\n    -   `date_last_evaluated` — when the submitter last reviewed the\n        classification\n\n## Typical Workflows\n\n### Count-First Workflow (Recommended)\n\nFor large or unknown result sets, use `count` first to decide whether to\nproceed, then `search` (which auto-paginates and returns `total_count` \u002F\n`fetched_count`), then `summary` to screen.\n\n```bash\n# Step 1: Gauge size (optional — search also returns total_count)\nuv run scripts\u002Fclinvar_api.py count \\\n  --query \"HBB[gene] AND pathogenic[clinsig]\" --output count.json\n\n# Step 2: Fetch all variant IDs (auto-paginates)\nuv run scripts\u002Fclinvar_api.py search \\\n  --query \"HBB[gene] AND pathogenic[clinsig]\" --output ids.json\n\n# Step 3: Get summaries (extract variant_ids from search output)\nuv run scripts\u002Fclinvar_api.py summary \\\n  --variant_ids 12345 67890 --output summary.json\n```\n\n### Deep Dive: search → evidence\n\nWhen you need the full clinical picture for a specific variant — including\nsubmitter rationales, PubMed citations, ontology-linked conditions, and allele\ncoordinates — use `evidence`.\n\n```bash\nuv run scripts\u002Fclinvar_api.py evidence \\\n  --variant_id 12345 --output evidence.json\n```\n\n### Workflow: Robust Variant Discovery (Triangulation)\n\nClinVar metadata is inconsistent. To fulfill \"List all\" requests, do not rely on\na single filter. Perform the following in a single turn and merge results:\n\n1.  **Search by exact label** (e.g., `\"3 prime UTR\n    variant\"[molecular_consequence]`).\n2.  **Search by HGVS nomenclature pattern** (e.g., `c.*`).\n3.  **Search by genomic coordinate range** (using `[chrpos]`).\n\nThis \"triangulation\" ensures structural variants with missing labels are not\noverlooked.\n\n### Verifying Coding vs. Non-Coding Status via HGVS\n\n`molecular_consequences` alone can be ambiguous (e.g., `splice donor variant`\nappears in both coding and non-coding contexts). Always cross-check the `title`\nfield for HGVS patterns:\n\n-   `c.-…` — 5' UTR (non-coding)\n-   `c.*…` — 3' UTR (non-coding)\n-   `c.123+N` \u002F `c.123-N` — intronic (non-coding)\n-   `p.Trp146Arg` etc. — protein effect (coding)\n\nA variant with UTR\u002Fintronic HGVS and no `p.` annotation is non-coding, even with\nsplicing labels. Conversely, any `p.` annotation indicates a coding effect.\n\n### ClinVar Metadata Reference\n\n-   **3' UTR**\n    -   Search String: `\"3 prime UTR variant\"[mol_consequence]`\n    -   HGVS: `c.*`\n-   **5' UTR**\n    -   Search String: `\"5 prime UTR variant\"[mol_consequence]`\n    -   HGVS: `c.-`\n-   To find \"high-confidence\" variants or expert-reviewed consensus, use the\n    `review_status` filter. This is the most efficient way to distinguish\n    between single-laboratory assertions and panel-reviewed ground truth.\n\n### When to Use Which Fields\n\n-   **Quick pathogenicity label** — Use `summary` → `clinical_significance`\n-   **Gene symbol and strand** — Use `summary` → `genes`\n-   **Variant type (SNV, del, etc.)** — Use `summary` → `variation_type`\n-   **Protein-level effect** — Use `summary` → `molecular_consequences`\n-   **Genomic coordinates (GRCh38)** — Use `evidence` → `allele_info`\n-   **Linked conditions (ontology)** — Use `evidence` → `conditions`\n-   **SO functional consequence** — Use `evidence` → `functional_consequences`\n-   **CNV breakpoints\u002Fcopy number** — Use `evidence` →\n    `structural_variant_details`\n-   **PubMed references** — Use `evidence` → `citation_references`\n-   **Date of last lab review** — Use both → `last_evaluated`\n-   **Clinician rationales** — Use `evidence` → `submissions[].curator_notes`\n\n### Retrieving Genomic Coordinates (Default HG38\u002FGRCh38)\n\nTo get precise genomic coordinates in the format `\u003Cchrom>:\u003Cpos>:\u003Cref>>\u003Calt>`\n(e.g., `chr5:70951945:G>A`), you must use the `evidence` command, as these\ndetails are not available in the `summary` output.\n\n**You MUST always include genomic coordinates in the format\n`\u003Cchrom>:\u003Cpos>:\u003Cref>>\u003Calt>` when listing or presenting variants, even if not\nexplicitly requested by the user. If coordinates are missing from the summary,\nuse the `evidence` command or dbSNP fallback to retrieve them.**\n\n1.  **Fetch Evidence**: Use `uv run scripts\u002Fclinvar_api.py evidence --variant_id\n    \u003CID> --output evidence.json`.\n2.  **Extract VCF Attributes**: The `evidence` command parses the XML. Extract:\n    *   Chromosome: `Chr`\n    *   Position: `positionVCF` (or `start`)\n    *   Ref: `referenceAlleleVCF` (or `referenceAllele`)\n    *   Alt: `alternateAlleleVCF` (or `alternateAllele`) from the\n        `SequenceLocation` element with `Assembly=\"GRCh38\"`.\n\n**Fallback for Imprecise Coordinates (Gene Range):** ClinVar often returns the\nfull gene range for non-coding variants. If the extracted coordinates correspond\nto the gene range instead of a specific position, use the `dbsnp-database` skill\nto resolve the precise coordinates using the `dbsnp_rsid` or HGVS title: 1.Check\nfor `dbsnp_rsid` in the `evidence` output. 2. Run `uv run scripts\u002Fdbsnp_cli.py\nresolve-rsid {rsid}` to get precise GRCh38 coordinates. 3. Format as\n`\u003Cchrom>:\u003Cpos>:\u003Cref>>\u003Calt>` using the SPDI or HGVS data from dbSNP.\n\n### Structural Variant Note\n\nThe `structural_variant_details` field is **only populated for copy number\nvariants (CNVs)**. For standard SNVs and small indels this field will be `null`.\nUse the `allele_info` fields (`position_start`, `position_stop`,\n`reference_allele`, `alternate_allele`) instead.\n\n### CNV \u002F Large Deletion Note\n\nLarge copy-number variants (CNVs) frequently have empty\n`molecular_consequences`. If a variant title mentions \"del\" and coordinates\noverlap your target region, it is relevant regardless of missing labels.\n\n### Obtaining and Using an API Key\n\nYou can register for a key for free at\nhttps:\u002F\u002Fwww.ncbi.nlm.nih.gov\u002Faccount\u002Fsettings\u002F. You **MUST** use the safe\ncredentials protocol in the `credentials` skill to check for and request this\nkey if this skill looks relevant to the user's request.\n\n## Best Practices\n\n-   Always use `uv run` to execute `python`.\n-   If `jq` is unavailable pivot immediately to using Python one-liners for\n    processing JSON (e.g., `uv run python3 -c \"import json; ...\"`).\n-   Use `count` before `search` to understand the result set size.\n-   The `search` command fetches all results by default and includes\n    `total_count` and `fetched_count` in the output — always verify these match\n    to confirm complete retrieval.\n-   Entrez results are **unsorted**. To order by date, fetch all results and\n    sort locally by `last_evaluated`.\n\n## Common Mistakes\n\n-   **Attempting to parse the E-utilities XML yourself** — Always use the\n    provided `clinvar_api.py` client which handles the unpredictable XML schemas\n    robustly.\n-   **Getting HTTP 429 Too Many Requests** — The client throws an exception\n    telling you to pause. You **MUST** use the safe credentials protocol in the\n    `credentials` skill to check for and request the `NCBI_API_KEY` to help the\n    user add it to their `.env` file, then retry.\n-   **Sending raw DNA sequences to the API** — The API expects HGVS\n    nomenclature, RS IDs, or proper Entrez coordinate syntax (`11[chr] AND\n    1234[chrpos]`), not raw ATCG strings.\n-   **For synonymous or non-coding variants** — HGVS nomenclature (e.g., CAPN3\n    AND \"c.551C>T\") is more reliable than coordinate searches ([chrpos]), as\n    many ClinVar records for these types lack precise genomic mappings.\n-   **Case sensitivity in molecular consequences** — ClinVar returns mixed-case\n    strings. Always use case-insensitive matching (`.lower()`) when filtering.\n-   **Parsing `search` output as a bare list** — `search` returns a JSON object\n    with `total_count`, `fetched_count`, and `variant_ids` — not a bare list.\n",{"name":5,"description":7},"main","42203fde5eb692ba254e058168ee1af75855fcb4","archive\u002Fbioinformatics\u002Fclinvar_database",0,"https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills","skills.marvellofaisal.me","https:\u002F\u002Fgithub.com\u002FMarvellbrazil","marvellofaisal.me","Software Engineering Student | Backend & API Developer | Tech Enthusiast | Homelabber","2026-10-23 22:23:52",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add Marvellbrazil\u002Fmy-agentic-skills --skill clinvar-database",{"tool":40,"command":41},"bunx","bunx skills add Marvellbrazil\u002Fmy-agentic-skills --skill clinvar-database",{"tool":43,"command":44},"pnpm","pnpm dlx skills add Marvellbrazil\u002Fmy-agentic-skills --skill clinvar-database","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Farchive\u002Fbioinformatics\u002Fclinvar_database\u002FSKILL.md",true,[48,54],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"references\u002Fcitation.bib","REFERENCE","参考文档",1098,"https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Farchive\u002Fbioinformatics\u002Fclinvar_database\u002Freferences\u002Fcitation.bib",{"path":55,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":58,"sourceHtmlUrl":59},"scripts\u002Fclinvar_api.py","SCRIPT","脚本",23185,"https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Farchive\u002Fbioinformatics\u002Fclinvar_database\u002Fscripts\u002Fclinvar_api.py",[61,68,75,82,89,96],{"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},"01M7DDQP2F4E42AMTGY0VCJ6G2","writing-guidelines","Review docs\u002Fprose for Writing Guidelines compliance. Use when asked to \"review my docs\", \"check writing style\", \"audit prose\", \"review docs voice and tone\", or \"check this page against the writing handbook\".","https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-guidelines\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},"01M7DDQM00XGSGT8V797C89R2E","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-for-agents\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},"01M7DDQEF1HP9CMH88HMN9MCBA","wrangler","Cloudflare Workers CLI for deploying, developing, and managing Workers, KV, R2, D1, Vectorize, Hyperdrive, Workers AI, Containers, Queues, Workflows, Pipelines, and Secrets Store. Load before running wrangler commands to ensure correct syntax and best practices. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwrangler\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},"01M7DDQA32ZAG7J4PMRXRSQ3NQ","workers-best-practices","Reviews and authors Cloudflare Workers code against production best practices. Load when writing new Workers, reviewing Worker code, configuring wrangler.jsonc, or checking for common Workers anti-patterns (streaming, floating promises, global state, secrets, bindings, observability). Biases towards retrieval from Cloudflare docs over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Fskills\u002Fworkers-best-practices\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},"01M7DDQ5482QSC0FC4NX107FVY","web-perf","Analyzes web performance using Chrome DevTools MCP. Measures Core Web Vitals (LCP, INP, CLS) and supplementary metrics (FCP, TBT, Speed Index), identifies render-blocking resources, network dependency chains, layout shifts, caching issues, and accessibility gaps. Use when asked to audit, profile, debug, or optimize page load performance, Lighthouse scores, or site speed. Biases towards retrieval from current documentation over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Fskills\u002Fweb-perf\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":19},"01M7DDQ1GZ7Z2Y67460SJJEMC0","using-agent-skills","Discovers and invokes agent skills. Use when starting a session, or when you need to decide which skill or workflow applies to the piece of work at hand. This is the meta-skill that governs how all other skills are discovered and invoked.","https:\u002F\u002Fgithub.com\u002FMarvellbrazil\u002Fmy-agentic-skills\u002Fblob\u002Fmain\u002Fskills\u002Fusing-agent-skills\u002FSKILL.md",[],[]]