agkan-add
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill frequently executes shell commands using the
agkanCLI and thermutility. It passes user-provided strings for titles, bodies, and IDs as arguments to these commands. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection and command injection through user-controlled data.
- Ingestion points: User input for fields like 'Title', 'Body', 'Tags', 'Priority', and 'Branch' collected via
AskUserQuestioninSKILL.md. - Boundary markers: None. While the instructions suggest wrapping arguments in double quotes (e.g.,
agkan task add "<title>"), there are no instructions for the agent to escape shell metacharacters (like backticks, dollar signs, or double quotes) within those inputs. - Capability inventory: The skill allows file writing (creating
/tmp/task_body.md), file deletion (rm), and execution of theagkantask management CLI. - Sanitization: There is no evidence of sanitization, validation, or escaping of external content before it is interpolated into shell command strings.
Audit Metadata