The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files (e.g., existing PRDs, HLDs, and configuration files) to understand project context. Maliciously crafted content within these files could attempt to influence the agent's behavior or output.
Ingestion points: SKILL.md instructs the agent to scan the filesystem using Glob patterns and read content from identified files.
Boundary markers: The skill does not define strict delimiters or "ignore instructions" warnings for the content read from files before processing.
Capability inventory: The agent has access to WebSearch, Glob, and AskUserQuestion tools.
Sanitization: No explicit sanitization or filtering of the content extracted from project files is performed before it is used to influence the drafting process.
[SAFE]: The skill implements security best practices by explicitly excluding sensitive directories such as .git/, node_modules/, and .venv/ from its file scanning operations to prevent accidental exposure of sensitive metadata or dependencies.
[SAFE]: The skill incorporates a human-in-the-loop mechanism by requiring user confirmation via the AskUserQuestion tool before reading the contents of the scanned project files, providing a layer of oversight.

prd-writer