The Agent Skills Directory

[DATA_EXPOSURE]: The skill reads local configuration files at ~/.pipa/communication-style.md and ~/.pipa/CONNECTORS.md. These files are used to determine presentation styles and tool preferences for project management integrations. Access is scoped to the skill's operational requirements.
[COMMAND_EXECUTION]: The skill utilizes composio-mcp for discovery and interaction with external project management tools (Jira, Linear, Slack, etc.). The instructions strictly enforce a 'read-only by default' policy and explicitly require human approval before any external write or update operation is performed.
[PROMPT_INJECTION]: The skill processes data from external sources, including Slack messages, emails, and issue comments, which constitutes an indirect prompt injection surface. However, the instructions mitigate this by directing the agent to categorize data as 'source facts' versus 'inferred signals' and focusing on status synthesis rather than command execution from processed content.

pipa-status-update