The Agent Skills Directory

[PROMPT_INJECTION]: No patterns attempting to override agent behavior, bypass safety filters, or extract system prompts were detected in the instructions.
[DATA_EXFILTRATION]: No hardcoded credentials, sensitive file path access, or unauthorized network transmission logic was found.
[EXTERNAL_DOWNLOADS]: The skill's metadata contains an installation command that fetches content from a GitHub repository. As GitHub is a well-known service, this is considered a standard distribution method and does not present a security risk.
[COMMAND_EXECUTION]: The skill provides documentation for Mercurial (hg) commands. These are legitimate tools for version control and are consistent with the skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent on how to clone external repositories via hg clone. While this represents a potential ingestion point for untrusted data from remote sources, the skill itself provides no mechanisms to execute that data unsafely or interpolate it into prompts without boundaries.

mercurial-hg