ripgrep
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to search and process content from the local filesystem, which exposes the agent to instructions embedded in untrusted data. 1. Ingestion points: The agent ingests data through search results from files in the repository (SKILL.md). 2. Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions when reading results. 3. Capability inventory: The skill utilizes powerful tools including Bash, Write, and Read (SKILL.md). 4. Sanitization: No methods are provided for sanitizing or filtering file content before the agent processes it.
- [COMMAND_EXECUTION]: The skill provides guidance on executing shell commands via the Bash tool for complex text searches and file modifications.
- Evidence: The instructions detail the use of shell pipelines like 'rg -l | xargs sed' for in-place editing (SKILL.md) and the '--pre' flag for running preprocessor commands on files (references/cli_reference.md).
Audit Metadata