Skills
Navigate
skills/f-labs-io/agent-html-skills/html-skills-stop/Gen Agent Trust Hub

html-skills-stop

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled shell script (scripts/stop.sh) to identify and terminate process IDs associated with the session.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads a MONITOR_ID from a temporary file in /tmp and uses this value to instruct the agent to execute the TaskStop tool. This represents an indirect prompt injection surface.
  • Ingestion points: The script reads from /tmp/html-skills-$SID.monitor-id and echoes the content to the agent's context via the shell output.
  • Boundary markers: There are no explicit boundary markers or validation steps to ensure the MONITOR_ID is legitimate before it is passed to the TaskStop tool.
  • Capability inventory: The skill utilizes the TaskStop agent tool and the shell kill command.
  • Sanitization: No sanitization or verification of the file content is performed before usage.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 12:48 AM