The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates an extensive surface for indirect prompt injection attacks by processing data from untrusted external websites.
Ingestion points: The skill facilitates reading content from over 79 platforms, including social media feeds (Twitter, Reddit, Facebook), search results, and arbitrary URLs via opencli web read (as documented in SKILL.md and commands.md).
Boundary markers: The instructions lack boundary markers or explicit guidance for the agent to distinguish between tool instructions and data retrieved from external sites.
Capability inventory: The agent is encouraged to execute shell commands (passthrough to gh, docker, lark-cli), post to social media (opencli twitter post), and modify local source files (the "Self-Repair" section in SKILL.md instructs the agent to fix adapter source code files).
Sanitization: There is no mention of sanitization or validation of the external content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill instructs the user to install an external, third-party package @jackwener/opencli from the npm registry to function. While this is the core tool of the skill, it introduces a dependency on external code maintained by a third party.
[COMMAND_EXECUTION]: The skill is designed to execute shell commands using the OpenCLI tool. This includes a "passthrough" feature that allows the agent to run subcommands for other powerful CLI tools such as GitHub (gh), Docker, and Vercel, extending the agent's execution capabilities.

qiaomu-opencli-usage