The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's workflow documentation explicitly instructs the agent to execute a shell command (node .gitnexus/run.cjs analyze) in the terminal under specific conditions (when the index is stale).
[DYNAMIC_EXECUTION]: The execution targets a script file (.gitnexus/run.cjs) located within the target repository being explored. This allows the execution of arbitrary code provided by the repository itself, which may not be trusted.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, potentially untrusted codebases through the gitnexus:// resource scheme.
Ingestion points: Data enters the agent context through resources like gitnexus://repo/{name}/context, gitnexus://repo/{name}/clusters, and gitnexus://repo/{name}/process/{name}.
Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions embedded within the codebase being analyzed.
Capability inventory: The skill documentation suggests the capability to execute terminal commands (node .gitnexus/run.cjs analyze) and perform deep-dive analysis on symbols via the context and query tools.
Sanitization: There is no mention of sanitizing the content retrieved from the repository before processing it or using it to trigger actions.

gitnexus-exploring