soia-pkm-reading-plan
Warn
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill defines optional dependencies installed from external repositories. While one dependency is from a well-known service, another is sourced from a third-party user repository (alchaincyf/huashu-weread).
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its data processing workflow.
- Ingestion points: The skill reads book lists and articles from user-provided text or URLs (documented in SKILL.md).
- Boundary markers: There are no explicit instructions to the agent to ignore embedded commands within the ingested text.
- Capability inventory: The skill has the ability to write files to the local Obsidian vault and interact with external APIs.
- Sanitization: The skill does not describe any validation or sanitization of the content extracted from external sources.
- [COMMAND_EXECUTION]: The skill interacts with the local system and external services.
- It reads from and writes to the local filesystem using the path defined in the OBSIDIAN_VAULT environment variable.
- It performs network operations to fetch book metadata from WeChat Reading APIs.
Audit Metadata