The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data (audit reports) from external specialists or user inputs to generate task descriptions in a project management system. \n- Ingestion points: Audit reports in Markdown or JSON format are ingested from specialist agents or directly from user input as described in SKILL.md and templates/agent-integration.md. \n- Boundary markers: The skill lacks explicit boundary markers or instructions to treat ingested report content as untrusted data, which could allow malicious instructions embedded in a report to influence agent behavior. \n- Capability inventory: The skill utilizes Teamwork MCP tools (e.g., mcp__teamwork__twprojects-create_task, mcp__teamwork__twprojects-create_milestone) to write data to an external project management platform. \n- Sanitization: There is no evidence of sanitization or validation of the ingested findings before they are interpolated into task templates.

teamwork-exporter