build-mobile

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill implements a telemetry system that reads configuration from ~/.superstack/config.json. If enabled, it transmits usage metadata including skill name, execution phase, event status, and platform architecture to a remote URL defined in the configuration via curl. It also maintains a local log file at ~/.superstack/telemetry.jsonl. This represents standard telemetry collection for a developer tool.\n- [EXTERNAL_DOWNLOADS]: The skill suggests cloning an external repository, https://github.com/sendaifun/solana-app-kit.git, for rapid prototyping. This repository is not owned by the skill author or a trusted vendor.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as its workflow relies on state read from .superstack/build-context.md.\n
  • Ingestion points: Reads from .superstack/build-context.md in SKILL.md.\n
  • Boundary markers: None present; the skill treats the file content as authoritative instructions for stack selection.\n
  • Capability inventory: Shell command execution (bash), package installation (npm), and repository cloning (git).\n
  • Sanitization: No validation or escaping is performed on the data retrieved from the build context file.\n- [COMMAND_EXECUTION]: The skill's Preamble and Telemetry sections execute shell commands to manage environment setup, directory creation, and log usage data to the filesystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 11:05 AM
Security Audit — agent-trust-hub — build-mobile