The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local utility scripts using python3, specifically {vault}/07-脚本与工具/surface_candidates.py for retrieving topic candidates and scripts/rebuild_published_index.py for indexing published content. These are functional components of the skill's data management workflow.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect injection as it ingests data from external local sources including {vault}/08-技能记忆/, _排行_summary.csv, and 知识库/索引.csv.
Ingestion points: Data enters the context from CSV files and local memory/vault directories.
Boundary markers: The skill relies on structured data formats (CSV/Markdown tables) but lacks explicit prompts to the agent to ignore instructions embedded within these data sources.
Capability inventory: The agent has the capability to execute shell commands (python3) which could potentially be abused if ingested data contained malicious parameters, though no such logic is currently present.
Sanitization: No explicit sanitization or filtering of the ingested file content is described in the instructions.
[SAFE]: References to external documentation and updates are directed to the author's own repository (github.com/iamzifei/aikey), which is consistent with the vendor identity and used for legitimate informational purposes.

aikey-topic