The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard shell utilities including git log, grep, wc, and ls to collect repository signals such as commit history and file counts. These are routine read-only operations used for state assessment.
[DATA_EXPOSURE]: The skill contains a hardcoded absolute directory path (/Users/kameronkales/planfi-app) in the decision workflow arguments. While this exposes the author's local environment structure, it does not represent a malicious exfiltration risk.
[INDIRECT_PROMPT_INJECTION]: The skill ingests content from local project files—specifically strategy.md, journal.md, and backlog.md—to inform its decision-making logic. This creates a surface where instructions embedded in those files could influence agent behavior.
Ingestion points: ceo/strategy.md, ceo/journal.md, and ceo/backlog.md are read and passed to the ceo-prioritize workflow.
Boundary markers: Absent; the content of these files is interpolated directly into the workflow arguments without explicit delimiters.
Capability inventory: The skill can execute shell commands, write to the local file system (ceo/journal.md), and invoke other specialized skills via the Skill tool.
Sanitization: No explicit sanitization or filtering of the ingested file content is performed prior to processing.

ceo