The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The scripts/init_skill.py script downloads Markdown documentation and an index file from https://www.ricequant.com/doc/. These resources are used to build a local searchable index for the agent. The source is the official domain of the skill author (ricequant).
[COMMAND_EXECUTION]: The skill requires the execution of local Python scripts such as init_skill.py to prepare documentation and inspect_env.py to verify the environment. These are standard utility scripts for managing the skill's dependencies and cache.
[CREDENTIALS_UNSAFE]: The scripts/rqamsc_runtime.py script accesses sensitive information, including a username and password, from a local configuration file (typically config.json in the user's application data directory). This is standard behavior for an SDK-based tool to facilitate authentication with the Ricequant AMS service, following common CLI configuration patterns.
[INDIRECT_PROMPT_INJECTION]: The skill ingests remote Markdown documentation and provides it as context to the agent. This creates a potential surface for indirect prompt injection if the vendor's documentation site were compromised, as the agent relies on this content to generate responses. The impact is minimized by the skill's specific focus on technical API usage.

rqamsc-python