skill-issue
Warn
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing a global npm package '@misha_misha/skill-issue' and a Homebrew tap 'mishanefedov/skill-issue/skill-issue' from a community member not on the trusted vendor list.\n- [REMOTE_CODE_EXECUTION]: Instructions encourage using 'npx @misha_misha/skill-issue', which executes code directly from a remote public registry.\n- [COMMAND_EXECUTION]: The skill provides commands to run the tool on local paths (e.g., '~/.claude/skills'), allowing it to read metadata and modify files if the '--fix' flag is used.\n- [PROMPT_INJECTION]: The auditing function creates an indirect prompt injection surface when processing external metadata. \n
- Ingestion points: Reads SKILL.md files from user-specified directories.\n
- Boundary markers: None; the skill does not instruct the agent to ignore commands embedded in audited files.\n
- Capability inventory: File reading, console output, and file modification (via '--fix').\n
- Sanitization: No validation or sanitization of ingested content is mentioned.
Audit Metadata