Back to skill

Security audit

Deprecated - AgentPMT AgentAddress + Paid Marketplace

Security checks for vulnerabilities and agentic risk

Overview

This is a simple index skill that points users to AgentPMT skill pages and install commands, with no bundled executable code or hidden behavior found.

Before installing the referenced AgentPMT skills, confirm the ClawHub page starts with the listed AgentPMT publisher URL and check that the resolved package/version matches what you intended, since the example commands install by current slug rather than immutable digest.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:42
Finding
Unpinned Remote Skill Installation Creates Supply-Chain Risk## Vulnerability Details **File Location**: `SKILL.md:42-62` **Vulnerability Type**: Unpinned third-party skill dependencies **Risk Level**: Medium ```markdown | Skill | AgentPMT ClawHub Page | Install Command | |---|---|---| | AgentPMT no-account AgentAddress/x402 setup | https://clawhub.ai/agentpmt/agentpmt-no-account-agentaddress-x402 | `openclaw skills install agentpmt-no-account-agentaddress-x402` | | What Is AgentPMT | https://clawhub.ai/agentpmt/what-is-agentpmt | `openclaw skills install what-is-agentpmt` | | AgentPMT account MCP/REST API setup | https://clawhub.ai/agentpmt/agentpmt-account-mcp-rest-api-setup | `openclaw skills install agentpmt-account-mcp-rest-api-setup` | ```bash openclaw skills install agentpmt-no-account-agentaddress-x402 openclaw skills install what-is-agentpmt openclaw skills install agentpmt-account-mcp-rest-api-setup ``` Install the current AgentPMT product skill published on the org account: | Skill | AgentPMT ClawHub Page | Install Command | |---|---|---| | Image Generation Agent | https://clawhub.ai/agentpmt/image-generation-agent | `openclaw skills install image-generation-agent` | ```bash openclaw skills install image-generation-agent ``` ``` ### Technical Analysis The documented commands install third-party skills using mutable slugs without specifying immutable versions or artifact digests. Although `SKILL.md` lists expected versions and canonical publisher pages elsewhere, the installation commands do not enforce those versions or verify the integrity of the retrieved artifacts. Consequently, the effective content installed by these commands can differ from the content that existed when this index was reviewed. The referenced remote skill packages are not included in the audited project, so their instructions, scripts, permissions, and runtime behavior could not be assessed. A compromised publisher account, malicious future release, registry-resolution failure, or ...[truncated 1565 chars]
Remediation
## Remediation Suggestions 1. Pin every dependency to an exact immutable version in each installation command, where supported. 2. Prefer cryptographic artifact digests or signed release identifiers over version labels alone. 3. Use fully qualified publisher and package identifiers to reduce ambiguity in registry resolution. 4. Verify publisher signatures, checksums, and package provenance before installation. 5. Retrieve and review each remote skill's instructions, scripts, declared permissions, network behavior, and transitive dependencies before activation. 6. Execute newly installed skills in a least-privilege sandbox with restricted filesystem, credential, network, and tool access. 7. Maintain an approved lockfile or allowlist recording the exact reviewed artifact for each skill. 8. Fail closed if the requested immutable release is unavailable or if integrity verification fails; do not silently install the latest release. 9. Update the documented commands so the versions shown in the catalog are actually enforced during installation.
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Static analysis

No suspicious patterns detected.