The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the agent to execute a local JavaScript file (node scripts/validate-wiki.mjs) to validate changes (Step 11). This is a functional requirement for wiki maintenance but involves running repository-provided code.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface (Category 8) because it transforms potentially untrusted data into persistent documents. (1) Ingestion points: User-provided conversation history and external technological guidance fetched from the web. (2) Boundary markers: Absent; there are no instructions to ignore or delimit potentially malicious content within the source material. (3) Capability inventory: The skill allows file creation and modification in the docs/wiki directory and command execution via node and git. (4) Sanitization: There are no directives to sanitize or validate the content extracted from conversations before it is written to the filesystem.

to-wiki