Firm Advanced Security Pack

Advanced security audit pack covering secrets lifecycle, path canonicalization, exec plan freeze, hook routing, config includes, prototype pollution, safeBin...

Security Scan

The skill claims to be an advanced OpenClaw audit pack and lists eight audit commands, but it provides no code, no install instructions for those tools, and no provenance for the declared dependency — the runtime expectations are under-specified and potentially misleading.

Purpose & Capability

The name, description, and listed tools are coherent with an OpenClaw configuration audit pack. However, the SKILL.md lists eight executable tool names (openclaw_*) yet the skill declares no required binaries and provides no installation or provenance for those executables. The metadata lists a dependency on 'mcp-openclaw-extensions >= 3.0.0' but the skill does not explain how that dependency supplies the tools or how it will be installed.

Instruction Scope

Runtime instructions direct the agent to run specific commands (e.g., openclaw_secrets_lifecycle_check config_path=/path/to/config.json). Those commands would execute arbitrary code on the host if present; the SKILL.md does not say where the commands come from, how to verify them, or any sandboxing/validation. The only file path referenced is a user-supplied config_path, which is reasonable for an audit tool, but the agent is being asked to run external commands without provenance—this is scope creep relative to an instruction-only skill.

Install Mechanism

No install spec is present (low disk-write risk), which is consistent with an instruction-only skill. However, the declared dependency (mcp-openclaw-extensions >= 3.0.0) is not accompanied by an installation or verification mechanism; the skill neither declares required binaries nor how to obtain the listed tools. That gap creates uncertainty about how the tools are expected to appear on the system.

Credentials

The skill requests no environment variables, no credentials, and no config paths. That is proportionate. Note: the commands it tells the agent to run will likely read user-supplied config files (config_path) — expected for an audit tool, but the skill does not constrain or document what parts of configs are read or transmitted.

Persistence & Privilege

always is false and the skill is user-invocable; it does not request permanent presence or elevated platform privileges. Autonomous invocation is allowed by default but is not combined with other high-risk flags here.

What to consider before installing

This skill is instruction-only and lists eight audit commands but provides no code, no install steps, and no provenance for the dependency it references. Before installing or running it: 1) Ask the publisher where the openclaw_* tools come from and how to obtain and verify 'mcp-openclaw-extensions >= 3.0.0' (signed releases, repository, package registry). 2) Do not run the listed commands on production systems until you can inspect the actual binaries/scripts; run them in a sandbox. 3) Request human-reviewed source code or a trusted install mechanism; if none is provided, treat the skill as untrusted because it could cause arbitrary command execution. 4) If you proceed, verify digitally-signed packages or review the extension code to ensure the tools only read the intended config files and do not exfiltrate secrets.

⚠️ Contenu généré par IA — validation humaine requise avant utilisation.

Purpose

Deep security auditing for OpenClaw configurations — covers external secrets lifecycle, channel path canonicalization, execution plan freeze validation, hook session routing, $include directive guards, prototype pollution detection, safeBins profile enforcement, and group policy default audit.

Tools (8)

Tool	Description	Severity
openclaw_secrets_lifecycle_check	External Secrets lifecycle audit	CRITICAL
openclaw_channel_auth_canon_check	Channel path canonicalization	CRITICAL
openclaw_exec_approval_freeze_check	Exec plan freeze validation	CRITICAL
openclaw_hook_session_routing_check	Hook session routing audit	HIGH
openclaw_config_include_check	$include directive guards	HIGH
openclaw_config_prototype_check	Prototype pollution detection	HIGH
openclaw_safe_bins_profile_check	safeBins profile enforcement	HIGH
openclaw_group_policy_default_check	Group policy default audit	HIGH

Usage

skills:
  - firm-advanced-security-pack

# Run full advanced security audit:
openclaw_secrets_lifecycle_check config_path=/path/to/config.json
openclaw_config_prototype_check config_path=/path/to/config.json
openclaw_safe_bins_profile_check config_path=/path/to/config.json

Requirements

• mcp-openclaw-extensions >= 3.0.0