The Agent Skills Directory

[SAFE]: No malicious behavior, obfuscation, or unauthorized data access patterns were detected. The skill's functionality aligns with its stated purpose of providing a CLI interface for platform management.
[COMMAND_EXECUTION]: The skill executes the dc command-line tool. It follows a defensive 'search-inspect-execute' protocol defined in SKILL.md and references/catalog.md, which requires the agent to verify command schemas and authentication status before running operations.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes data from external sources (platform messages and audit logs).
Ingestion points: Data enters the context via commands like dc global-management message list-messages and dc global-management audit list-audits found in references/modules/global-management.md.
Boundary markers: The skill emphasizes the use of structured JSON output (-o json) and explicit command inspection (dc commands show) to help the agent distinguish between data and instructions.
Capability inventory: The skill can perform sensitive administrative tasks, such as managing users, roles, SSH keys, and access tokens.
Sanitization: There is no explicit instruction within the skill to sanitize the content of messages or logs before processing, though the structured workflow mitigates accidental execution risks.

dc