Index of every page on this site: llms.txt
All skills
k-dense-ai avatar

/genomic-intelligence

@ec59a29
Run it yourself

Predicts regulatory features, gene structure, and expression directly from DNA sequence using Genomic Intelligence's hosted transformer DNA language models — no local GPU or model weights. Six tasks over a REST API and a hosted MCP server (keyless public demo): promoter regions, splice donor/acceptor sites, enhancer activity, chromatin state, sequence-to-expression (log TPM), and de-novo gene annotation, plus a composite find-genes-then-predict-expression workflow. Use when the user has a gene symbol, a genomic region, or a DNA/FASTA sequence and wants any of these predictions, mentions Genomic Intelligence, genomicintelligence.ai, api.genomicintelligence.ai, or mcp.genomicintelligence.ai.

Use this Skill: https://skilld.dev/gh/k-dense-ai/scientific-agent-skills/genomic-intelligence

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesreview.md

≈1.1k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Contract review — 2026-10-01

Reviewed all five original skill/reference files against the deployed OpenAPI, current official guides, live MCP discovery, and released gi-mcp source. The served API identifies itself as 2026.09.22.2 (af902d84) with x-contract-revision: 16; the released local adapter is 0.1.0a21.

Primary sources

  • Live OpenAPI: all 12 operations, auth exceptions, literal task requests, options, response schemas, job status/limit semantics, extension fields, and VCF boundary.
  • REST guide: transport, text-format restrictions, input/output envelopes and contract revision.
  • Task guide: interpretation, strand/token-span limitations, description sensitivity, composite behavior, and VCF workflow's unfinished scientific output.
  • Limits and errors: 20,000-site splice response cap, error-details exception, burst/rpm distinction, timeouts and retention. The splice cap is not structurally declared in the reviewed OpenAPI; do not infer its absence from the schema alone.
  • MCP guide, hosted https://mcp.genomicintelligence.ai/mcp self-description, and gi-mcp 0.1.0a21: acquisition envelopes, demo restrictions, cached/UCSC reference sources, canonical-TSS fallback, and fixed minus-strand window math. Published wheel SHA-256 was checked before source inspection.
  • Ensembl symbol lookup, ID lookup, and region sequence: public routes, species aliases, expanded transcripts and sequence bounds. Current indexed official docs were readable; direct requests failed below.

What was executed

  • Public /health, live schema and all six /v1/tasks/{task}/models calls returned HTTP 200; discovery output was checked against the served schema.

  • Empty-body unauthenticated POSTs to all eight task/workflow operations and GETs to both protected job endpoints returned HTTP 401 error envelopes. These verify auth behavior only, not predictions or input validation.

  • Hosted MCP initialize, tools/resources/prompts discovery, model lookup, four resource reads and three reference acquisition calls succeeded at the transport layer. Acquisition payloads confirmed data.ref, lengths and HBB minus-strand TSS geometry. Demo recent-jobs/resource and list-jobs behavior returned the documented unavailable error rather than any user's history.

  • Direct Ensembl HBB symbol lookup, stable-ID lookup and a 100-bp region call returned HTTP 500. Hosted HBB acquisition used catalog/cache; the small region used UCSC, so those successes do not establish Ensembl availability.

  • 24 local checks exercised the documented Python blocks with mocked HTTP, including all six URLs/bodies, expression context/index, bounded 202-to-200 polling, timeout/failed-job behavior, strand-dependent coordinate identity, and validation of recorded public responses.

  • Released adapter functions were exercised with synthetic reference bytes: both strand-dependent TSS windows, both canonical/gene-body fallback flags, and IUPAC reverse-complement behavior passed without upstream requests.

Limits of this review

No authenticated GI inference, free-demo inference, variant workflow, object storage write, model weights or biological benchmark was run. Biological outputs, timing claims, runtime 422s, text exports and paid tiers are supported by the current published contract/guides, not end-to-end inference evidence. MCP deployment and released adapter source are separate evidence; a version number alone cannot establish that every deployment behavior matches a wheel. Canonical transcript provenance and correct coordinate math do not validate an active tissue-specific TSS or a model's scientific accuracy.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning6d3 checks · Risk SAFE
  • Gen Agent Trust Hub6d

    The skill is safe for its intended purpose of DNA sequence analysis using hosted models. It correctly manages API keys through environment variables and uses standard libraries for networking. The verdict is low because the skill processes genomic data from external sources and user input, which creates a surface for indirect prompt injection, although the risk is mitigated by specific genomic alphabet validation.

  • Socket26d

    No alerts

  • Snyk26d

    Risk: MEDIUM · 1 issue

Signed by skilld at ec59a29. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 4 days ago.

Activeupdated 2 months ago
Other metadata
compatibility
Python 3.10+ with the `requests` library for the REST examples. Network access required. The REST `/v1` API needs a `GI_API_KEY` (a `gi_` bearer); the hosted MCP server at mcp.genomicintelligence.ai/mcp works keyless against a rate- and concurrency-limited public demo tier, key optional.
metadata
{
  "version": "1.4",
  "last-reviewed": "2026-10-01",
  "skill-author": "Genomic Intelligence",
  "trigger-keywords": "DNA sequence prediction, regulatory genomics, promoter prediction, splice site prediction, enhancer activity, chromatin state, gene expression prediction, sequence to expression, log TPM, gene annotation, transcript prediction, DNA language model, genomic intelligence, hosted inference, Ensembl sequence, FASTA prediction, cis-regulatory, TSS window, DeepSEA, DeepSTARR, BigBird splice, MCP genomics",
  "openclaw": {
    "primaryEnv": "GI_API_KEY",
    "envVars": [
      {
        "name": "GI_API_KEY",
        "required": false,
        "description": "Optional gi_ bearer key for the REST /v1 API and higher MCP rate and concurrency limits. The hosted MCP demo runs keyless; request a key at contact@genomicintelligence.ai."
      }
    ]
  }
}

README badge

README badge for k-dense-ai/scientific-agent-skills/genomic-intelligence