compose-gen
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the
--addcommand-line argument. The content of this argument is directly interpolated into the LLM prompt without any sanitization or boundary markers.\n - Ingestion points: The
extraparameter in thegenerateComposefunction (src/index.ts), which receives input from the-aor--addCLI options.\n - Boundary markers: Absent. The input is concatenated directly into the prompt string without delimiters or instructions to ignore embedded commands.\n
- Capability inventory: The skill has file system write capabilities via
fs.writeFileSyncinsrc/cli.ts(used to output the generated YAML file).\n - Sanitization: No sanitization, escaping, or validation is performed on the user-provided string before it is interpolated into the LLM prompt.
Audit Metadata