The Agent Skills Directory

[SAFE]: The skill primarily consists of YAML-based workflow templates, project structures, and reporting configurations. No malicious patterns such as obfuscation, command execution, or unauthorized data access were found.
[INDIRECT_PROMPT_INJECTION]: The skill defines automation rules that ingest untrusted data from external sources, specifically through GitHub integrations (e.g., {{issue.title}}, {{issue.body}}) and intake forms.
Ingestion points: GitHub issue/PR sync rules and internal request forms in SKILL.md.
Boundary markers: None used during interpolation of external variables into task descriptions or Slack messages.
Capability inventory: Uses MCP tools asana_create_task and asana_update_task to write data based on these inputs.
Sanitization: No explicit sanitization or validation of the external content is defined. While this represents a surface for indirect prompt injection, it is considered a functional requirement for the skill's primary purpose of automation.

Asana Automation