The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (e.g., scripts/content/analyze_transcript.py, scripts/content/extract_themes.py) and shell utilities like find and ls to process data. These operations are restricted to the local environment and used for intended analysis purposes.
[DATA_EXPOSURE_AND_EXFILTRATION]: Data access is scoped to the references/ and output/ directories. No network-enabled tools (e.g., curl, wget) are requested in allowed-tools, and no external communication patterns were identified in the instructions.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a processing engine for untrusted external data (transcripts and documents), creating a potential surface for indirect injection.
Ingestion points: Reference files in references/transcripts/, references/examples/, and references/competitors/.
Boundary markers: None explicitly defined in the provided markdown instructions.
Capability inventory: The agent has access to Bash, Read, Write, and the ability to execute the skill's Python scripts.
Sanitization: No specific input sanitization or filtering logic is described in the prompt structure.

content-analyzer