The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill contains a template for an automated 'auditor subagent' designed to scan repositories for outdated TUI patterns. This creates a potential surface for indirect prompt injection where malicious content in the audited source files could attempt to influence the agent's behavior.\n
Ingestion points: The subagent is instructed to read and grep source code from the target repository path provided in references/audit-prompt.md.\n
Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions within the source files being analyzed.\n
Capability inventory: The audit template includes capabilities to read file contents and write a markdown report back to the repository (docs/ratatui-modernization.md).\n
Sanitization: No sanitization or validation logic is provided for the content ingested from the external repository.\n- [EXTERNAL_DOWNLOADS]: The skill uses curl to fetch the current version of the ratatui crate from crates.io and provides multiple links to official project documentation on ratatui.rs and docs.rs. These are well-known and reputable services used for legitimate developer workflows.

ratatui-patterns