The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The hermes_check.py script retrieves versioning information from the official GitHub API (api.github.com/repos/NousResearch/hermes-agent) to check for updates. Additionally, getting-started-tutorials.md references an installation script hosted on raw.githubusercontent.com.
[COMMAND_EXECUTION]: The skill frequently executes the local hermes CLI binary using Python's subprocess module. These executions are used to perform health checks, retrieve configuration paths, and generate command help summaries for the agent's use.
[DATA_EXFILTRATION]: While the skill accesses sensitive files such as .env and auth.json to verify configuration, it includes explicit redaction logic in hermes_ops_check.py and hermes_slack_check.py that masks API keys and Slack tokens before they are printed or incorporated into the agent's context.
[PROMPT_INJECTION]: The skill uses scripts/hermes_check.py to ingest output from hermes --help and generate a local command map (command-map.generated.md). This creates an indirect prompt injection surface where the agent's behavior could be influenced by instructions embedded in local command documentation, although the risk is minimal in this context.

hermes