The Agent Skills Directory

[SAFE]: The skill implements a benign business workflow for process automation. It retrieves metrics from a local file and updates a configuration state based on defined criteria or user confirmation. No malicious patterns, obfuscation, or unauthorized data access behaviors were detected.
[NO_CODE]: The skill is comprised solely of markdown instructions and does not include any scripts, binaries, or other executable files.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from a local file, creating a potential surface for injection if the file content is compromised. 1. Ingestion points: Reads metrics from blueprint/60-bandeja.md. 2. Boundary markers: None specified. 3. Capability inventory: Updates configuration state 'paso 3' to 'automatico'. 4. Sanitization: None present.

soltar