seqkit
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill enables processing of external FASTA and FASTQ files through various seqkit subcommands, creating a vulnerability surface where malicious instructions embedded in sequence headers could influence the agent's subsequent actions. Ingestion points: FASTA/FASTQ files processed by subcommands like grep, replace, and fx2tab as described in SKILL.md and references/recipes.md. Boundary markers: Absent; no specific instructions are provided for the agent to use delimiters or ignore embedded instructions when reading tool output back into its context. Capability inventory: The skill utilizes shell command execution (seqkit, awk, sed, printf) and file system writes via shell redirection. Sanitization: Absent; the skill does not mention sanitizing sequence data before processing it or reading the output.
- [COMMAND_EXECUTION]: The skill provides numerous recipes and instructions for the agent to execute shell commands, pipes, and process substitutions using seqkit, awk, cut, and sed on the local filesystem. This functionality is the primary purpose of the skill.
Audit Metadata