The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, potentially untrusted data to perform its core functions.
Ingestion points: The skill ingests user-provided email drafts for critique, performance data (JSON) for sequence analysis, and context files like .claude/product-marketing-context.md.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from being influenced by instructions embedded within the user-provided drafts or external data.
Capability inventory: The agent has the capability to write and modify email content, and to execute a local analysis script (scripts/email_sequence_analyzer.py) via a subprocess.
Sanitization: The skill does not implement validation, sanitization, or filtering for the data ingested from the user or the environment before processing it.
[COMMAND_EXECUTION]: The skill uses a local Python utility for sequence analysis.
The instructions specify the use of python3 scripts/email_sequence_analyzer.py to analyze JSON formatted email sequences. This is a standard execution of a local tool included with the skill to provide quality metrics such as word counts, reading level, and spam trigger detection.

cold-email