experiment-code

Warn

Audited by Runlayer on Feb 22, 2026

Risk Level: MEDIUM
Scan Summary
Max Score
78%
Files
3
Flagged
3
Chunks
4
Flagged Files (3)
SKILL.mdHIGH
78.3%

Malicious tool definition detected

Tool: SKILL.md Description: --- name: experiment-code description: Write ML experiment code with iterative improvement.

references/code-patterns.mdHIGH
78.3%

Malicious tool definition detected

Tool: references/code-patterns.md Description: # Experiment Code Patterns Reference ## Pattern 1: Experiment Execution Loop (AI-Scientist) ```python MAX_ITERS = 4 # Max fix attempts per run MAX_RUNS = 5 # Max experiment runs MAX_STDERR_OUTPUT = 1500 # Truncate stderr def perform_experiments(idea, folder_name, coder, baseline_results): current_iter = 0 run = 1 next_prompt = initial_prompt.format(...) while run < MAX_RUNS + 1: if current_iter >= MAX_ITERS: break coder_out = coder.run(n

references/experiment-prompts.mdHIGH
78.3%

Malicious tool definition detected

Tool: references/experiment-prompts.md [1/2] Description: # Experiment Code Prompts Reference Extracted from AI-Scientist, AgentLaboratory, and AI-Researcher. ## 1. Initial Experiment Prompt (AI-Scientist) ``` Your goal is to implement the following idea: {title}.

Tool: references/experiment-prompts.md [2/2]

Audit Metadata
Max File Score
78%
Classification
UNKNOWN_SERVER
Files Scanned
3
Files Flagged
3
Chunks Analyzed
4
Analyzed
Feb 22, 2026, 06:30 AM
Security Audit — runlayer — experiment-code