The Agent Skills Directory

[SAFE] (SAFE): The skill consists entirely of markdown documentation and code examples for Elixir performance auditing. No malicious behavior was detected.\n- [Prompt Injection] (SAFE): No instructions to override system prompts or bypass safety filters were found. The instructional text is consistent with the stated purpose of code review.\n- [Data Exposure & Exfiltration] (SAFE): There are no hardcoded credentials, sensitive file paths, or network operations targeting non-whitelisted domains. References to external tools in code snippets (e.g., HTTPClient) are illustrative and not executed by the agent.\n- [Unverifiable Dependencies & Remote Code Execution] (SAFE): No package managers (pip, npm, hex) are invoked, and no remote scripts are downloaded or executed. The skill links to a local relative path (../review-verification-protocol/SKILL.md), which is standard for multi-skill repositories.\n- [Indirect Prompt Injection] (LOW): While the skill's primary function is to process untrusted data (Elixir source code), there is no evidence of exploitable capabilities within the skill files (such as exec, eval, or file-write operations) that could be triggered by malicious code being reviewed.

elixir-performance-review