feature-forge
Fail
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: CRITICALEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The documentation URL provided in the skill (https://jeffallan.github.io/claude-skills/skills/workflow/feature-forge/) has been flagged as malicious and blacklisted by automated scanners (URLite). Furthermore, the core skill file (SKILL.md) has been flagged by file reputation scanners (FileRepMalware [Misc]), indicating potential associations with malicious content or low-reputation software distribution.
- [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting and processing untrusted data from both users and external codebases to generate specifications.
- Ingestion points: Data is gathered from users through the
AskUserQuestionstool and technical findings are retrieved from the codebase usingTasksubagents (e.g.,architecture-designer,security-reviewer, andExploretasks). - Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from mistaking data within the codebase or user responses as new system-level instructions.
- Capability inventory: The skill possesses the ability to write files to the local file system (
specs/directory), execute theAskUserQuestionstool, and spawn additional autonomous subagents (Task). - Sanitization: The instructions lack requirements for sanitizing, escaping, or validating the content retrieved from external subagents or user input before it is interpolated into the final specification templates.
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