ai-slop-detection
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of instructional text providing heuristics for detecting AI-generated content. It does not perform any automated actions or utilize external tools.
- [NO_CODE]: There are no scripts, binaries, or configuration files provided that would execute on the host system. All logic resides within the natural language instructions provided to the agent.
- [PROMPT_INJECTION]: The instructions include a directive to "not trust pre-existing reports." While this influences agent behavior, it is a standard instruction for analysis tasks and does not attempt to bypass safety filters or override system constraints.
- [DATA_EXPOSURE]: The skill mentions specific URLs and domains associated with AI platforms (e.g., bolt.new, lovable.dev) as search patterns. These are used for identification purposes within the user's project and do not involve exfiltrating data or accessing credentials.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data (the contents of a user project). While this presents a theoretical surface for indirect prompt injection if an attacker includes malicious instructions in project files, the skill itself lacks any dangerous capabilities (like file writing or network access) that could be leveraged in an exploit chain.
Audit Metadata