ccg
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of global NPM packages
@openai/codexand@google/gemini-cli. These packages are associated with established organizations (OpenAI and Google). - [COMMAND_EXECUTION]: The workflow relies on executing shell commands via the
omc askCLI tool. It passes dynamically generated strings as arguments to these commands, which is the primary mechanism for model orchestration. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8). It ingests untrusted data from external sources and has the following characteristics:
- Ingestion points: Reads model output artifacts from
.omc/artifacts/ask/(SKILL.md). - Boundary markers: None identified; instructions do not specify the use of delimiters or warnings to ignore embedded commands in advisor outputs.
- Capability inventory: Uses the Bash tool to execute
omc askcommands (SKILL.md). - Sanitization: No sanitization or validation of the advisor-generated markdown files is performed before synthesis.
Audit Metadata