The Agent Skills Directory

[COMMAND_EXECUTION]: The skill extensively uses the Bash tool to perform file operations, including searching for entries, validating file headers, and performing date-based logic using grep, awk, and date utilities. These commands are primarily used to maintain the content-calendar.md file as the single source of truth.
[DYNAMIC_EXECUTION]: The instructions direct the agent to execute specific local Python scripts: scripts/video-transcript.py (which takes a user-supplied URL as an argument) and ~/private-skills/scripts/scan-negative-framing.py. Additionally, python3 -c is used for inline date arithmetic. These executions rely on the presence of local environment scripts which are not included in the skill package itself.
[INDIRECT_PROMPT_INJECTION]: The skill demonstrates a vulnerability surface for indirect prompt injection by processing external and potentially untrusted data.
Ingestion points: The agent reads the content-calendar.md file (which may be manually modified), processes user-provided articles/transcripts, and ingests content from external video URLs.
Boundary markers: The instructions lack explicit requirements for using boundary markers or delimiters when interpolating external content into the model's context.
Capability inventory: The skill has access to Bash and Write tools, enabling it to execute shell commands and modify local files based on its instructions.
Sanitization: There is no mention of sanitizing, escaping, or validating the external content before it is used to influence the agent's content creation or pipeline management tasks.

content-calendar