io.github.athril/agentic-target-evidence

What it mainly works on, as our AI model read its tool descriptions: nothing outside itself: it only answers questions.

None of its 42 tools are described as changing anything. For 2 the reading was uncertain, so no mark is shown.

• chembl_get_chemistryreads or searches

  “Fetch ChEMBL drug-mechanism and bioactivity signals for a ChEMBL target id.”

  2 inputs, 1 required

  • chembl_target_idstringrequired
  • gene_symbolstring
• clingen_get_validityreads or searches

  “Fetch ClinGen gene-disease validity classifications for a gene.”

  1 input, 1 required

  • gene_symbolstringrequired
• clinicaltrials_search_trialsreads or searches

  “Search ClinicalTrials.gov v2 API for studies involving the gene and disease.”

  3 inputs, 2 required

  • genestringrequired
  • diseasestringrequired
  • populationstring or null
• clinicaltrials_count_condition_trialsreads or searches

  “Count trials for a disease/condition regardless of target gene or intervention.”

  1 input, 1 required

  • conditionstringrequired
• depmap_get_dependencyreads or searches

  “Fetch DepMap CRISPR gene effect and dependency scores for a gene.”

  1 input, 1 required

  • gene_symbolstringrequired
• dgidb_get_gene_drug_interactionsreads or searches

  “Fetch curated drug-gene interaction claims for a gene from DGIdb.”

  3 inputs, 1 required

  • gene_symbolstringrequired
  • max_resultsinteger
  • approved_onlyboolean
• dgidb_get_gene_categoriesreads or searches

  “Fetch DGIdb gene-category annotations for a gene.”

  1 input, 1 required

  • gene_symbolstringrequired
• encode_get_regulatory_coveragereads or searches

  “Fetch regulatory-assay (ChIP-seq/DNase-seq/ATAC-seq) coverage at a gene locus.”

  2 inputs, 1 required

  • gene_symbolstringrequired
  • genomestring
• expression_atlas_get_differential_expressionreads or searches

  “Fetch disease-vs-control differential expression for a gene from Expression Atlas.”

  4 inputs, 1 required

  • gene_symbolstringrequired
  • diseasestring
  • speciesstring
  • ensembl_idstring
• gencc_get_validityreads or searches

  “Fetch GenCC's per-submitter gene-disease validity classifications for a gene.”

  1 input, 1 required

  • gene_symbolstringrequired
• gnomad_get_constraintreads or searches

  “Fetch gnomAD gene-level LoF/missense/synonymous constraint (LOEUF, pLI, pRec, MOEUF, syn_z).”

  1 input, 1 required

  • gene_symbolstringrequired
• gnomad_get_clinvar_variantsreads or searches

  “Fetch ClinVar variants overlapping this gene from gnomAD's integrated dataset.”

  2 inputs, 1 required

  • ensembl_idstringrequired
  • gene_symbolstring

Each quoted sentence is the first sentence of the publisher’s own description. The kind of tool and the “changes data” mark are our AI model’s reading of that description, not something we tested: two differently worded runs had to agree, otherwise no label is shown, and nobody has checked them by hand. A tool without the “changes data” mark is described as only reading, searching or calculating, or the two runs did not agree. No tool was called. Read the publisher’s full description before relying on it. Inputs are the parameters in the tool's own input schema: name, declared type, whether it is required, and the first sentence of its description. Read 2026-10-23.

The registry verifies a package once, when it is published, and does not re-check it. A difference below is about the listing, not a judgement of the publisher.