The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The maintenance script scripts/check-updates.py performs a network request to the npm registry (registry.npmjs.org) to verify package versions. This is an administrative function targeting a well-known service and is considered safe under standard maintenance procedures.
[EXTERNAL_DOWNLOADS]: The skill documents the use of an official Upstash utility (https://upstash.com/start-redis) which allows for the dynamic creation of ephemeral Redis databases for development purposes.
[SAFE]: Analysis of all 12 reference files confirms they contain legitimate documentation and instructional code snippets for the Upstash platform. All external URLs and package dependencies point to official vendor domains or trusted public registries.
[SAFE]: The skill promotes secure credential management by instructing users and agents to use environment variable loaders (e.g., Redis.fromEnv()) rather than hardcoding sensitive API tokens.
[INDIRECT_PROMPT_INJECTION]: The skill describes an attack surface for indirect prompt injection as it facilitates processing data from external databases and message queues. However, the documentation mitigates this by providing patterns for signature verification (in QStash) and data serialization best practices. Ingestion points include Redis keys and Vector metadata; capability inventory includes data retrieval and task scheduling; boundary markers are managed via SDK-level serialization.

upstash