heating the forge...

@forgecat/anthropics_skills_algorithmic-art — Security Risk Report | ForgeCat

anthropics_skills_algorithmic-art

Security risk report for @forgecat/anthropics_skills_algorithmic-art v0.0.4

Low

Profile sourced from Anthropic's official skills repository (github.com/anthropics/skills).
Declared author and repository are legitimate and verifiable; no supply-chain red flags.

SRC-010High

Content instructs the AI to repeatedly emphasize that outputs should appear 'meticulously crafted,' 'refined through countless iterations,' and 'the product of deep expertise by someone at the absolute top of their field' — this is guidance poisoning that systematically biases future outputs toward false claims of human authorship and expertise.
The instruction 'Emphasize craftsmanship REPEATEDLY' and 'This framing is essential - repeat phrases like...' explicitly directs the AI to inject deceptive language into generated art philosophies, misleading users about the origin and quality of AI-generated work.
The 'ESSENTIAL PRINCIPLE' section instructs embedding 'subtle, niche reference[s]' as 'conceptual DNA woven invisibly into parameters' — directing the AI to hide intentional design choices and present them as emergent, which is deceptive guidance about the nature of the generated output.

Details

Findings

Evidence

'Emphasize craftsmanship REPEATEDLY: The philosophy MUST stress multiple times that the final algorithm should appear as though it took countless hours to develop, was refined with care, and comes from someone at the absolute top of their field.'
'repeat phrases like "meticulously crafted algorithm," "the product of deep computational expertise," "painstaking optimization," "master-level implementation."'
'The concept is a subtle, niche reference embedded within the algorithm itself... The reference must be so refined that it enhances the work's depth without announcing itself.'

Low

No tools or skills are declared in the profile.
No file_write, shell, web_fetch, or other high/medium-risk permissions are requested.
The profile is purely instructional guidance for the AI's reasoning process, not a tool definition.

Low