get-task
Fail
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute
taskmd get $ARGUMENTSwhere$ARGUMENTSis the raw user query. This is a command injection vulnerability because shell metacharacters (e.g.,;,&&,$()) in the user input are not escaped or sanitized, allowing an attacker to execute arbitrary system commands. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. 1. Ingestion points: The
Readtool is used to load task file contents into the agent's context. 2. Boundary markers: There are no delimiters or instructions to ignore embedded commands within the ingested task files. 3. Capability inventory: The agent has access to theBashtool, which can be exploited by malicious instructions hidden in a task file. 4. Sanitization: There is no evidence of validation or filtering for the content read from the task files.
Recommendations
- AI detected serious security threats
Audit Metadata