plan-ceo-review
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands via the Bash tool to perform a 'Pre-Review System Audit'. The commands include git log, git diff, and grep to collect repository history and metadata. It also uses shell commands to create directories and write review files to the local file system.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes content from local project files and web search results that are not controlled by the skill itself.\n
- Ingestion points: The skill reads local files (CLAUDE.md, AGENTS.md, TODOS.md), project source files containing markers like TODO or FIXME, and results from the WebSearch tool.\n
- Boundary markers: No explicit delimiters or instructions are used to separate this ingested data from the skill's primary instructions.\n
- Capability inventory: The skill utilizes the Bash, Read, Grep, and WebSearch tools to interact with and process data.\n
- Sanitization: Content from the files and search results is processed directly without sanitization or validation.
- [EXTERNAL_DOWNLOADS]: The skill uses the WebSearch tool to conduct a 'Landscape Check', fetching external market research and competitor data from the internet into the agent's context.
Audit Metadata