embedded-systems

Fail

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: CRITICALEXTERNAL_DOWNLOADSMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes an external documentation link at https://jeffallan.github.io/claude-skills/skills/specialized/embedded-systems/ which has been flagged as malicious and blacklisted by automated scanners.
  • [METADATA_POISONING]: The skill's main file, SKILL.md, has been identified by reputation scanners as malicious (FileRepMalware). Additionally, there is a discrepancy in author attribution; the metadata points to an external GitHub profile (Jeffallan) associated with the flagged infrastructure, rather than the expected author (emmraan).
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by ingesting untrusted project requirements and hardware constraints to generate firmware code.
  • Ingestion points: User-provided specifications for MCU types, memory limits, and timing requirements described in the 'Core Workflow'.
  • Boundary markers: None; the skill lacks explicit instructions for the agent to ignore potentially malicious directions embedded in hardware datasheets or user-provided requirements.
  • Capability inventory: The skill possesses capabilities to write code that interacts with low-level system peripherals (GPIO, I2C, SPI, CAN) and manages real-time tasks, which could be exploited if malicious instructions are processed.
  • Sanitization: There are no verification or sanitization steps defined for hardware constraints or user inputs before they are interpolated into the implementation logic.
Recommendations
  • CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
  • AI detected serious security threats
  • Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Sep 6, 2026, 11:36 AM
Security Audit — agent-trust-hub — embedded-systems