The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill includes a configuration command npx skills add Gusen1453/creed in SKILL.md for adding a suite of companion skills. This fetches additional instruction sets from the author's GitHub repository, which is a standard distribution method for this ecosystem.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user inputs via a 'Minimal task prompt template' defined in SKILL.md. The ingestion points include task intent and constraint fields. Boundary markers are implemented through specific header labels in the template. The skill's capability inventory is restricted to logic routing and repository exploration, with no direct involvement in high-risk operations like file system modification or network exfiltration. No specific sanitization or filtering of the user-provided content is defined within the skill instructions.

using-creed