The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to load and execute instructions from a external task file located at .aiox-core/development/tasks/dev-develop-story.md. This creates an indirect prompt injection surface where a malicious or compromised task file could provide instructions that override agent behavior or safety constraints.
Ingestion points: Reads the content of a file path provided in $ARGUMENTS[0] and a specific task file in the .aiox-core directory.
Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded prompts within the loaded task files.
Capability inventory: The skill is authorized to perform file system modifications, manage git branches, and execute shell commands.
Sanitization: There is no evidence of validation or sanitization of the content loaded from the external task files before the agent executes them.
[COMMAND_EXECUTION]: The skill explicitly executes shell commands including npm run lint, npm run typecheck, and npm test. While these are standard development tasks, they represent a command execution vector that depends on the integrity of the local package.json and the project environment.

develop-story