agent-swarm-deployer
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands for data management tasks. Specifically, it uses
wc -lto count total items and thesplitutility to partition large CSV files into smaller batches for sub-agents. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection. Ingestion points: Untrusted data enters the agent context through file reads, directory globbing, and CSV parsing (SKILL.md, Step 1a, 4b). Boundary markers: While the skill uses structured JSON for result aggregation, the briefs sent to sub-agents lack explicit delimiters or instructions to ignore embedded commands within the processed data (SKILL.md, Step 3). Capability inventory: The skill utilizes
Bash,Write, andAgenttools across its scripts and sub-agent deployments (SKILL.md frontmatter). Sanitization: There is no evidence of sanitization, validation, or escaping of external content before it is interpolated into sub-agent instructions.
Audit Metadata