The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill identifies and processes untrusted external data, including user-provided bug reports, error logs, and stack traces, which could be used to deliver malicious instructions to the agent.
Ingestion points: The 'Core Process' in SKILL.md instructs the agent to capture observed behavior and gather evidence from logs and error messages during the triage steps.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the ingested data to prevent the agent from following instructions embedded within the logs.
Capability inventory: The skill is designed for an environment where the agent likely has filesystem access (to write reproduction specs like spec/requests/orders_spec.rb) and command execution capabilities (to run bundle exec rspec as described in the 'Canonical Request-Boundary Example').
Sanitization: The instructions do not describe any sanitization, validation, or escaping of the ingested logs or reports before the agent processes them or generates shell commands.

triage-bug