The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from issue tracker bodies and comments, which could be used to influence the agent's triage decisions.
Ingestion points: The agent is instructed to read issue metadata, bodies, and comments from an external tracker to classify and mutate tickets.
Boundary markers: The instructions do not define specific delimiters or provide guidance to the agent on ignoring instructions embedded within the ticket text.
Capability inventory: The skill has the authority to execute local scripts and perform tracker mutations (updating labels, statuses, and body content) via MCP tools.
Sanitization: No explicit sanitization or validation logic is defined for handling text retrieved from the issue tracker.
[COMMAND_EXECUTION]: The skill relies on the execution of multiple Node.js scripts via the shell.
Evidence: The instructions explicitly direct the agent to run ../ziw-orchestrate/scripts/tick-snapshot.mjs, ../ziw-orchestrate/scripts/tick-plan.mjs, and ../ziw-orchestrate/scripts/linear-dag-start.mjs using the node command.
Context: These scripts are located in a sibling directory (ziw-orchestrate), indicating a functional dependency on external project components. The skill assumes these scripts are safe and available in the execution environment.

ziw-triage