The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent fetches and analyzes external documentation URLs provided by the user (Step 1 and Step 2) to derive implementation steps, configuration requirements, and code samples for new skills.
Ingestion points: External documentation URLs provided by the user for SAP Help, GitHub, or API references (SKILL.md, Step 1).
Boundary markers: The skill does not provide specific delimiters or instructions to ignore embedded prompts within the fetched documentation content.
Capability inventory: The skill performs file system writes (creating and updating SKILL.md and reference files) and executes shell commands for validation (find, wc, grep).
Sanitization: No explicit sanitization or validation logic is defined to prevent the model from obeying instructions that might be maliciously embedded in the documentation links.
[COMMAND_EXECUTION]: The skill includes shell commands intended for the agent or user to verify the structure and content of the generated skill files.
Evidence: Commands such as wc -l, find, head, and grep are used for validating line counts, verifying file existence, and checking frontmatter parsing in the final validation section.
[EXTERNAL_DOWNLOADS]: The skill workflow involves fetching content from external documentation sites to inform the skill creation process.
Evidence: Step 1 and Step 2 explicitly instruct the agent to fetch and analyze documentation from sources like SAP Help and GitHub repositories. These are well-known and authoritative services for the skill's intended purpose.

create-skill-sap-fiori-tools