The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple shell commands to gather diagnostic information, including curl for API connectivity checks, dig for DNS lookups, openssl for TLS verification, and tar for bundle creation. It also uses npm list and pip show to verify SDK installations.
[DATA_EXPOSURE]: The skill accesses sensitive local files, specifically .env and application logs (*.log). It attempts to mitigate the risk of accidental exposure by implementing automated redaction using sed, which masks all environment variable values and ElevenLabs API key patterns (sk_*) before they are written to the output bundle.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from local log files, which creates a surface where malicious instructions embedded in logs could theoretically influence the agent during the collection process.
Ingestion points: Ingests contents of *.log files in the current working directory (references/implementation.md).
Boundary markers: None; raw log lines matching the grep filter are appended to the error report.
Capability inventory: The skill has access to shell execution, network probing via curl, and package management tools (node, pip, npm) as specified in the implementation script.
Sanitization: Includes automated redaction of API keys and environment values using sed patterns.

elevenlabs-debug-bundle