The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external files to perform requirement analysis and scoring.
Ingestion points: Reads docs/01-需求调研/确认需求清单.md, 需求调研报告.md, and 市场调研报告.md.
Capability inventory: The skill uses Read, Write, AskUserQuestion, Agent, and Bash tools.
Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings when passing file content into subagent prompts.
Sanitization: There is no evidence of sanitization, escaping, or validation of the external file content before it is interpolated into subagent tasks.
[COMMAND_EXECUTION]: The preamble executes a shell script located relative to the skill directory.
Evidence: bash "$(dirname "${BASH_SOURCE[0]}")/../../check-update.sh" 2>/dev/null || true in the Preamble section. This is a common pattern for skill maintenance but represents a local command execution vector.

pm-priority