The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/generate_error_report.py contains a generate_script method that constructs bash scripts (.sh) by combining a predefined template with content provided via the --content command-line argument. The script then programmatically applies execution permissions (chmod 0o755) to the newly created file. This pattern represents a dynamic execution surface where malicious input could lead to the creation and execution of arbitrary code.\n- [PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data from application logs and metrics, which constitutes an indirect prompt injection surface.\n
Ingestion points: The skill is configured to read application logs and metrics (as seen in SKILL.md instructions and the allowed-tools including Read and Grep), while scripts/analyze_errors.py identifies files for analysis.\n
Boundary markers: There are no delimiters or explicit instructions provided to the agent to ignore or isolate instructions that may be embedded within the logs or metrics being processed.\n
Capability inventory: The skill has access to shell execution (Bash), file system operations, and the ability to generate executable scripts via scripts/generate_error_report.py.\n
Sanitization: No evidence of content validation or sanitization was found in the scripts that process external data or generate reports.

monitoring-error-rates