skills/freedomintelligence/openclaw-medical-skills/tooluniverse-immune-repertoire-analysis/Gen Agent Trust Hub
tooluniverse-immune-repertoire-analysis
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate bioinformatics analysis tools. No malicious code, obfuscation, or persistence mechanisms were detected. The scripts use established libraries like pandas, numpy, and scanpy for genomic data processing.
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection via ingestion of external data files (e.g.,
load_airr_datainSKILL.md). - Ingestion points: File reads in
load_airr_data(MixCR, 10x, AIRR formats). - Boundary markers: Absent.
- Capability inventory: Scientific tool calls via
tu.run_one_function(IEDB_search_tcells,PubMed_search) and Python data processing. - Sanitization: Absent.
- Note: The risk is minimal as the ingested data consists of biological sequences (TCR/BCR) which are processed as numeric or character strings without direct interpolation into command-like prompts.
Audit Metadata