The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill-card references the author's public GitHub repository (github.com/marioluciofjr/skills) for access to additional skills and information. This is a standard vendor-provided resource.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided instructions to generate structured behavioral templates.
Ingestion points: Processes user input in SKILL.md (Path B) to extract agent identity and rewrite instruction sets.
Boundary markers: The skill does not add specific delimiters or 'ignore' instructions to the generated instrucoes.txt file.
Capability inventory: The skill is restricted to text generation and file output for the user. It does not possess capabilities for subprocess execution, network operations, or sensitive file system access.
Sanitization: No sanitization or filtering of the input text is performed, as the skill is designed to faithfully restructure the user's provided agent identity.
[SAFE]: Extensive analysis for prompt injection, data exfiltration, obfuscation, and dynamic execution found no malicious patterns. The skill's behavior is consistent with its intended purpose as a development aid for AI identity design.

gerador-soul