The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill is instructed to access sensitive credentials by reading the ~/.config/catalyst-cloud/customer.json file to extract an authentication key.
[DATA_EXFILTRATION]: The skill transmits the extracted authentication key via an Authorization: Bearer header during curl network operations to a dynamic baseUrl defined in the configuration file.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from an external issue tracking system, which creates a vulnerability surface for indirect prompt injection.
Ingestion points: External ticket content and project activity data are retrieved from API endpoints including /api/v1/issues and /api/v1/search.
Boundary markers: The instructions do not define any delimiters or provide warnings to the agent to ignore instructions embedded within ticket descriptions or comments.
Capability inventory: The skill possesses the capability to perform network operations (curl) and modify the state of tickets in the tenant.
Sanitization: There is no mention of sanitization, validation, or escaping of the ingested ticket content before it is processed by the agent.

steward