skills/garrytan/gstack/skillify/Gen Agent Trust Hub

skillify

Warn

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes multiple local binaries (e.g., gstack-config, gstack-update-check, gstack-slug) located in ~/.claude/skills/gstack/bin/ to manage environment state, sessions, and telemetry logging.\n- [REMOTE_CODE_EXECUTION]: The primary function involves dynamically synthesizing TypeScript code (script.ts) and tests (script.test.ts) from the conversation history, which are then executed using the Bun runtime in a temporary directory for verification.\n- [DATA_EXFILTRATION]: It facilitates the transmission of usage telemetry to the vendor's tracking system and provides a mechanism to synchronize project-specific artifacts (such as plans and reviews) to a remote repository.\n- [EXTERNAL_DOWNLOADS]: The skill performs automated update checks via gstack-update-check and connects to remote git repositories for its synchronization functionality.\n- [PROMPT_INJECTION]: The instructions include directives that command the agent to treat the skill logic as executable instructions rather than reference material and to prioritize its specific behavioral patches over general system safety or planning guidelines.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 16, 2026, 06:25 PM
Security Audit — agent-trust-hub — skillify