The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides explicit shell commands in the SOP and startup sections for managing services using poetry, uvicorn, and nohup (e.g., poetry run uvicorn src.main:app --host 0.0.0.0 --port 8001).
[DATA_EXFILTRATION]: The skill documentation discloses absolute local system paths belonging to the developer (e.g., /Users/qitmac001395/workspace/QAL/ideas/apps/zhimeng-agent) and a specific hardcoded Feishu open_id (ou_18b8063b232cbdec73ea1541dfb74890) intended for automated message delivery.
[INDIRECT_PROMPT_INJECTION]: The skill automates the reading and extraction of content from Obsidian journal files to be sent to external platforms, which poses a vulnerability to malicious instructions embedded in those documents.
Ingestion points: Reads markdown files from local Obsidian Vault directories (~/Documents/Obsidian Vault/Journal/).
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the retrieved document content.
Capability inventory: The skill has capabilities to read files and invoke network-based messaging via the mcp__feishu__im_v1_message_create tool.
Sanitization: There is no evidence of content sanitization or validation before the extracted journal data is processed by the agent or sent to Feishu.

zhimeng-agent