The Agent Skills Directory

[PROMPT_INJECTION]: The skill operates by ingesting and processing arbitrary files from the user's repository, which creates an indirect prompt injection surface.
Ingestion points: The skill reads repository content using the Read, Glob, and Grep tools during the '探测文档布局' (Phase 0) and '派发撰写子代理' (Phase 2) stages as described in SKILL.md.
Boundary markers: No specific delimiters or instructions to ignore instructions embedded within the ingested code (e.g., within comments or docstrings) are implemented to protect the agent's logic.
Capability inventory: The skill has access to powerful tools including Write, Edit, Bash, and Agent as defined in the allowed-tools section of SKILL.md.
Sanitization: The instructions do not mention sanitizing or escaping the content retrieved from external files before it is passed to subagents or processed into documentation.
[COMMAND_EXECUTION]: The skill is configured with the Bash tool and interpolates user-provided input via the $ARGUMENTS variable in SKILL.md.
While the skill logic encourages using native tools for file exploration, the availability of the shell tool combined with direct argument injection provides a standard surface for unintended command execution.

docs-update