The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's preamble executes a local script check-update.sh using a relative path (../../check-update.sh). This assumes a specific repository structure and executes a script located outside the skill's own directory.
[COMMAND_EXECUTION]: Uses shell commands to create the docs/03-增长迭代 directory during the initialization phase.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection when processing strategy documents and user input into subagent tasks.
Ingestion points: Reads content from docs/05-产品战略/战略决策建议.md and takes user-provided business priorities.
Boundary markers: Absent. The subagent prompt template in Step 3 interpolates variable data ({O} and {产品现状/数据}) without using delimiters or clear separation between instructions and data.
Capability inventory: The skill possesses the ability to write to the filesystem (Write), execute shell commands (Bash), and delegate tasks to subagents (Task).
Sanitization: No input sanitization or content validation is performed on the ingested data before it is processed by the agent or its subagents.

pm-okr