The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local maintenance script scripts/version_check.sh during preflight initialization to verify that the skill and its associated toolkit are up to date.
[EXTERNAL_DOWNLOADS]: The update script performs a network request to the official GitHub API (api.github.com) to check for the latest release tags of the alva-ai/skills repository. Additionally, the skill's documentation directs the loading of standard libraries from well-known CDNs such as jsdelivr.net and unpkg.com for browser-side rendering.
[INDIRECT_PROMPT_INJECTION]: The skill provides an extensive framework for processing financial news, social media posts (X/Twitter, Reddit), and market sentiment. This represents an indirect injection surface.
Ingestion points: Data enters the context via the Arrays data skills and various search modules (references/search.md, references/data-skills.md).
Boundary markers: The skill explicitly requires the use of a 'VOICE' block in system prompts to ensure grounding and prevent the AI from adopting tones or instructions found in ingested snippets (references/user-facing-prose.md).
Capability inventory: The skill can perform HTTP requests, read/write to the Alva filesystem (ALFS), and execute trades through the platform's API.
Sanitization: Instructions mandate source attribution and warn against synthesizing verdicts from prompt-injected text, requiring that recommendations be driven by verified data pipelines rather than summaries.

alva