The Agent Skills Directory

[PROMPT_INJECTION]: The skill defines a workflow that processes documentation content, creating a surface for indirect prompt injection.
Ingestion points: The skill ingests content from markdown files in directories like docs/ and agent-docs/, as well as README.md and project memory files.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following malicious instructions hidden within the documentation content.
Capability inventory: The skill involves both reading project files and potentially modifying them when the --fix flag is enabled.
Sanitization: No sanitization, validation, or filtering of the input documentation content is specified.

enhance-docs