The Agent Skills Directory

[SAFE]: The skill accesses local configuration files within the ~/.pipa/ directory to determine communication styles and tool preferences. This is standard configuration management for the pipa toolset and does not involve accessing sensitive system credentials or protected OS files.
[PROMPT_INJECTION]: An indirect prompt injection surface exists because the skill is designed to process content from external sources that an attacker could potentially influence. • Ingestion points: The skill retrieves data from external tools including Linear, Google Drive, Notion, and Jira using the composio-mcp discovery mechanism. • Boundary markers: No specific delimiters or safety instructions are defined to separate the instructions of the skill from the content retrieved from external tools. • Capability inventory: The skill utilizes tool discovery and execution via composio-mcp, which includes the potential to perform external writes to the aforementioned platforms (mitigated by mandatory human approval gates). • Sanitization: There is no evidence of specific input validation, filtering, or sanitization performed on the content retrieved from external trackers or documents.

pipa-dependency-handoff