tldr-overview
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill executes local shell commands using the
tldrutility (e.g.,tldr tree,tldr structure). These operations are restricted to the local file system and are necessary for the skill's stated purpose of project analysis. - [DATA_EXPOSURE] (SAFE): While the skill reads project structure and function names, it does not contain any network exfiltration patterns or access to sensitive system credentials (like .ssh or .aws folders).
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection from analyzed code.
- Ingestion points: Files and directories within the target project (e.g.,
src/,.pyfiles). - Boundary markers: None; the output format directly interpolates tool output into the final report.
- Capability inventory: Shell command execution and file system reading.
- Sanitization: No sanitization or escaping is performed on the code structure or function names before they are presented to the agent. An attacker could name a function or file with malicious instructions to influence the agent's summary.
Audit Metadata