The Agent Skills Directory

[Hardened Network Communication]: The skill uses a custom HTTP opener that strictly enforces HTTPS for remote hosts and explicitly blocks HTTP redirects. This prevents potential credential leakage that could occur if a request were silently redirected to an untrusted domain.
[Secret Redaction System]: A comprehensive redaction mechanism is implemented using multiple regex patterns to mask sensitive tokens (e.g., Bearer tokens, API keys) in console output, logs, and error messages. This provides a defense against accidental disclosure of credentials in process transcripts.
[Secure Credential Loading]: The skill loads configuration from a dedicated environment file (~/.jira.env) using a custom parser rather than shell execution. It performs rigorous checks on file ownership, permissions, and file type (refusing symlinks) before reading, ensuring secrets are handled according to security best practices.
[Audit Logging]: A fail-closed audit logging system is included, which records request attempts and outcomes. The implementation ensures that if the audit log cannot be written, the operation is aborted, providing reliable traceability for enterprise environments.
[Data Processing Surface]: As a tool that fetches and displays content from an external Jira instance (such as issue summaries and comments), the skill acts as a conduit for untrusted data. While the data is redacted and structure is preserved, downstream components should treat this content as potentially containing instructions or malformed data that could influence agent behavior.

jira