slack-incident-workflow

Warn

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes shell commands using python3 to execute various automation tasks. It relies on the $MCP_CALL variable to perform these actions, which involves spawning subprocesses to handle network diagnostics and incident reporting.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes dynamic script paths determined by environment variables, including $PYATS_MCP_SCRIPT, $SERVICENOW_MCP_SCRIPT, and $GAIT_MCP_SCRIPT. This pattern constitutes dynamic code execution because the specific logic being run is not hardcoded but resolved at runtime from the system environment.
  • [PROMPT_INJECTION]: The skill possesses a vulnerability surface for Indirect Prompt Injection due to its reliance on external Slack data.
  • Ingestion points: The skill retrieves untrusted content from Slack using channels:history and groups:history scopes.
  • Boundary markers: There are no defined delimiters or instructions to help the agent distinguish between administrative commands and the text content of Slack messages being analyzed.
  • Capability inventory: The agent can execute system commands via $MCP_CALL, perform network pings, and interact with external systems like ServiceNow.
  • Sanitization: No input validation or sanitization mechanisms are described for the Slack message history before it is integrated into the agent's workflow.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 6, 2026, 12:32 AM