The Agent Skills Directory

[Indirect Prompt Injection Surface]: The skill facilitates the retrieval of secret data from Azure Key Vault (as seen in SKILL.md and references/examples.md). This creates a surface where the agent might process data that could contain untrusted instructions.
Ingestion points: Secret retrieval operations such as getSecret and listPropertiesOfSecrets in both documentation and example files.
Boundary markers: The instructions do not specify delimiters to isolate the retrieved secret values from other prompt content.
Capability inventory: The skill utilizes network access for Azure API communication and demonstrates local file writing for secret backups.
Sanitization: The skill does not include specific guidance for sanitizing or validating the content retrieved from the external vault before it is processed.
[Credential Placeholders in Documentation]: The skill's examples in SKILL.md and references/examples.md include placeholder strings like P@ssw0rd123! and sk_live_abc123xyz. These are identified as illustrative examples for SDK usage rather than actual hardcoded credentials and are standard in technical documentation.

azure-security-keyvault-secrets-java