ce-promote

Warn

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill encourages the installation and execution of a third-party Node.js package (@every-env/spiral-cli) from npm using npx to enable brand-voice matching features.
  • [COMMAND_EXECUTION]: The skill runs various shell commands including git, gh (GitHub CLI), and spiral to extract feature details from the project's history, PRs, and local configuration.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted content from external sources such as Pull Request bodies, Git diffs, and changelogs to derive marketing copy. This data could be manipulated to influence the agent's output.
  • [COMMAND_EXECUTION]: The skill programmatically modifies the local Git exclude file (info/exclude) to ensure that its local configuration files are not tracked by the version control system.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 23, 2026, 09:14 PM
Security Audit — agent-trust-hub — ce-promote