The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local shell commands for environment validation and setup. Specifically, it executes git rev-parse --is-inside-work-tree to verify repository status, git init to initialize new repositories, and mkdir -p to create directory structures. These operations are within the expected scope of a setup tool.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from productos/PLAN.md during its 'adoption' phase, which could serve as a vector for malicious instructions if the plan file is sourced from an untrusted party.
Ingestion points: The skill reads productos/PLAN.md from the filesystem in SKILL.md.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within the plan file during the verification pass.
Capability inventory: The skill has the ability to move files (mv), create directories (mkdir), copy or append to agent guidelines (CLAUDE.md/AGENTS.md), and update .gitignore.
Sanitization: There is no evidence of sanitization or filtering applied to the content of the plan file before it is processed or used to annotate the programme steps.

studio-setup