makecli

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to install the makecli utility using Homebrew from the author's repository (brew tap qfeius/makecli). This is the expected installation path for the vendor's tool.
  • [COMMAND_EXECUTION]: The skill uses the makecli command-line interface to perform resource management tasks such as creating apps, entities, and relations. It also includes a self-update functionality (makecli update) that retrieves updates from GitHub Releases.
  • [DATA_EXFILTRATION]: Configuration and credentials for the platform are stored locally in ~/.make/config and ~/.make/credentials. The skill correctly instructs the user to run interactive configuration commands (like makecli configure token) manually to ensure secure handling of sensitive tokens.
  • [PROMPT_INJECTION]: The skill processes local DSL YAML files and remote API responses to manage platform state. This introduces a surface for indirect instructions, although no specific malicious patterns were observed.
  • Ingestion points: makecli diff -f <path>, makecli apply -f <path>, and resource list commands in SKILL.md and references/cli-reference.md.
  • Boundary markers: None specified for the YAML or JSON data being processed.
  • Capability inventory: Shell command execution via makecli and local file system access in SKILL.md.
  • Sanitization: No explicit validation or escaping of external content is mentioned in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 01:18 PM