agent-reach
Warn
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches an installation guide from an external repository at
https://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/install.md. The instructions direct the agent to perform environment setup based on this remote file, which may involve executing unverified shell commands. - [COMMAND_EXECUTION]: The skill relies heavily on shell-based CLI tools including
agent-reach,mcporter,xreach,yt-dlp, and the GitHub CLI (gh) to search and interact with platforms. - [DYNAMIC_EXECUTION]: Several components use
python3 -cto execute inline Python code for processing platform-specific data, such as RSS feeds and WeChat articles. - [INDIRECT_PROMPT_INJECTION]: The skill has a significant attack surface for indirect prompt injection.
- Ingestion points: Processes content from 13+ external platforms including Twitter, Reddit, YouTube, XiaoHongShu, and general web searches.
- Capability inventory: The agent can execute shell commands, run Python code, and perform network requests.
- Boundary markers: Lacks explicit delimiters or instructions to ignore instructions within ingested data.
- Sanitization: Data retrieved from external platforms is processed without visible sanitization or validation.
Audit Metadata