ztp

A mandatory security audit skill for validating new code, skills, and MCP servers against the SEP-2026 Zero Trust protocol.

Before installing or trusting this as a Gatekeeper, consider the following: - Provenance: the skill has no homepage and an unknown source. Ask the publisher for provenance, a cryptographic release, or maintainer contact information before using it as an authoritative gate. - Path mismatch: SKILL.md shows a different script path (skills/openclawSecurity/...) than the included files (scripts/shield_pro.py). Confirm the correct runtime invocation and update the docs to avoid accidental failures. - Dynamic import/execution: the code references a dynamic scan harness (importlib/util and tests that mention a 'trap'). Dynamic importing can execute code; obtain and review the SafeImportHarness implementation to ensure it truly prevents side effects (no fork/exec, no uncontrolled os.system, no network during import). - External/semantic scanning: the tool can optionally call external tools or LLM-based semantic checks. Confirm whether any API keys or network endpoints would be used automatically, which environment variables they would read, and whether code or findings are transmitted outside your environment. - False positives and policy strictness: the forbidden-imports and forbidden-calls lists include commonly-used modules (e.g., functools, open). Expect false positives; plan for a manual-review workflow and test the tool on benign code to understand its rules. - Review risky constructs: review any code that uses importlib, eval/exec handling, or automated dynamic execution traps. Also run the included unit tests in an isolated sandbox to validate behavior. If you need to proceed: run the tool in an isolated environment (air-gapped or heavily restricted container), inspect the SafeImportHarness codepath, and verify that no network connections are made and no environment secrets are read before giving it gatekeeper status.

Type: OpenClaw Skill Name: ztp Version: 2.0.0 The OpenClaw AgentSkills bundle is a security audit tool designed to detect malicious patterns, obfuscation, and unauthorized network/system access in code and documentation. The `SKILL.md` provides clear, benign instructions for the AI agent to run security scans. The core `shield_pro.py` script implements static (AST, regex) and dynamic (SafeImportHarness) analysis to identify vulnerabilities and malware, but it does not contain any malicious logic itself. The `SafeImportHarness` actively prevents dangerous system calls during module import, acting as a defense mechanism. While `os.popen` is used in `generate_markdown_report` for a benign purpose (getting the current date), and `scan_semantic` checks for an `OPENCLAW_AI_API_KEY` without performing any external communication, these do not indicate malicious intent. The overall design and implementation align with its stated purpose as a security gatekeeper.