skills/egorfedorov/slot-casino-game-developer-skills-for-stake-engine/wasm-integration/Gen Agent Trust Hub
wasm-integration
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute a local Python script,
scripts/validate_wasm_bundle.py, which is used to perform sanity checks on build artifacts. The script operates on files specified in a user-provided manifest. - [DATA_EXPOSURE]: The validation script reads local files to check for WASM magic numbers and specific JavaScript strings. It does not include any network functionality, meaning that data read from these files cannot be exfiltrated by the skill itself.
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external project artifacts (WASM binaries and JS loaders). The analysis is limited to static string matching and binary header verification. The script does not execute the content of these files, nor does it interpolate their contents into instructions that could override the agent's behavior.
- [METADATA_POISONING]: The skill's metadata and documentation accurately reflect its functionality as a WASM integration and validation tool, with no deceptive instructions found in the frontmatter or reference files.
Audit Metadata