The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to run npx gitnexus, which fetches and executes code from the npm registry at runtime.
[COMMAND_EXECUTION]: The skill utilizes shell command execution for all its primary functions, such as npx gitnexus analyze, status, and wiki, which execute the downloaded package's logic on the local system.
[DATA_EXFILTRATION]: The wiki command includes a --gist flag that allows publishing repository documentation to a public GitHub Gist. This poses a risk of exposing proprietary information or sensitive codebase structures to the public internet.
[CREDENTIALS_UNSAFE]: The skill documents that LLM API keys are saved to ~/.gitnexus/config.json. This identifies a specific location for sensitive credentials that could be targeted for exposure.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection during the repository analysis phase.
Ingestion points: The analyze command parses all source files in the project root to build a knowledge graph.
Boundary markers: No boundary markers or instructions to ignore embedded instructions in source files are present.
Capability inventory: The skill can execute shell commands via npx and potentially perform network operations via the wiki tool's LLM integration and Gist publishing.
Sanitization: There is no evidence of sanitization or validation performed on the source code content before it is processed into the knowledge graph.

gitnexus-cli