The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to run the command taskmd verify $ARGUMENTS --format json. The variable $ARGUMENTS is derived directly from the user's input.
[PROMPT_INJECTION]: There is a risk of command injection because user-provided content is interpolated into a shell command string. A malicious user could provide a task ID containing shell metacharacters (e.g., ;, &&, |) to execute unauthorized commands.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through the data it processes.
Ingestion points: Codebase files and task descriptions read via Read, Glob, and Grep.
Boundary markers: Absent. The agent is instructed to read assertions and evaluate them without clear delimiters or instructions to ignore embedded malicious instructions.
Capability inventory: The agent has the ability to execute commands via Bash and access the filesystem using Read, Glob, and Grep.
Sanitization: No sanitization or validation of the file content is performed before the agent evaluates the assertions.

verify-task