The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the ghost-paper package from the npm registry. This package is the primary dependency for report generation and is provided by the skill author.
[REMOTE_CODE_EXECUTION]: By invoking npx, the skill executes code from a remote package repository (npm) to perform its core functions.
[COMMAND_EXECUTION]: Shell commands are employed to fetch guidelines via npx ghost-paper prompt and to compile reports using the npx ghost-paper build command.
[PROMPT_INJECTION]: The skill processes untrusted user data to generate reports, which creates a surface for indirect prompt injection.
Ingestion points: User-supplied data, metrics, and markdown drafts are used as the primary input for the report generation process (Step 2).
Boundary markers: There are no specified delimiters or 'ignore' instructions to isolate user-provided text from the structural markdown instructions.
Capability inventory: The skill allows the agent to write files to the local filesystem and execute shell commands through the npx utility.
Sanitization: Content provided by the user is written directly to the source markdown file without explicit sanitization or escaping of potential instructions.

ghost-paper