Back to skill

Security audit

Deliberate Practice For Manual Skills

Security checks for vulnerabilities and agentic risk

Overview

This skill is a plain coaching workflow for manual-skill practice, with the main risk being a disclosed but unpinned npx install command.

Before installing, prefer a pinned or otherwise trusted ClawHub installer version and run installation from a normal least-privileged account. Once installed, expect the skill to create or update practice logs and planning files for the manual skill you ask it to coach.

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:15
Finding

Unpinned Third-Party Package Execution Through npx

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 15–22
Vulnerability Type: T08: Insecure Dependencies
Risk Level: Medium

Vulnerable Code

yaml
openclaw:
  requires:
    tools: [filesystem]
  install: "npx clawhub install deliberate-practice-for-manual-skills"
shell
npx clawhub install deliberate-practice-for-manual-skills

Technical Analysis

The documented installation command invokes clawhub through npx without specifying an exact package version or integrity value. If the package is not already available locally, npx can retrieve it from the configured package registry and execute it with the permissions of the invoking user.

Because neither a version constraint nor a cryptographic checksum is supplied, the executable selected when the command is run may differ from the version that existed when this Skill was reviewed. The repository contains only SKILL.md; it does not include the retrieved package, a lockfile, or other information that would permit auditing the package implementation or its installation-time behavior.

This is a supply-chain trust issue rather than evidence that the current clawhub package is malicious. Exploitation depends on compromise or unauthorized modification of the package, its registry account, the configured package registry, or another relevant dependency-resolution mechanism.

Attack Path

  1. An attacker compromises the package publication account, registry entry, distribution channel, or another dependency-resolution component used for clawhub.
  2. The attacker publishes a modified package version containing malicious installation-time or runtime behavior.
  3. A user follows the instruction npx clawhub install deliberate-practice-for-manual-skills.
  4. npx resolves and downloads the mutable package version from the configured registry.
  5. The downloaded package executes under the invoking user's account.
  6. The malic ...[truncated 856 chars]
Remediation
View remediation

Remediation Suggestions

  1. Pin clawhub to a reviewed exact version, for example:
    shell
    npx --yes clawhub@<reviewed-exact-version> install deliberate-practice-for-manual-skills
    
  2. Verify the package archive against a trusted cryptographic digest or signed provenance record before execution.
  3. Document the expected registry explicitly and enforce a trusted registry configuration to reduce dependency-confusion and registry-redirection risks.
  4. Use a lockfile or an equivalent immutable dependency manifest covering the installer and its transitive dependencies.
  5. Review package lifecycle scripts and disable unnecessary installation scripts where the package manager supports doing so.
  6. Run installation in a sandbox or least-privileged environment without sensitive credentials, broad filesystem access, or administrative permissions.
  7. Prefer distributing an audited, versioned installer artifact with reproducible-build or provenance metadata rather than dynamically executing the latest registry release.
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Rp1

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

The skill instructs users to run npx clawhub install deliberate-practice-for-manual-skills without pinning a specific package version. Because npx resolves and executes code from the registry at install time, a later compromised, malicious, or breaking release of clawhub could be pulled and run automatically. The surrounding skill content is benign, but that does not reduce the supply-chain risk introduced by an unpinned executable installer command.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

This second occurrence repeats an unpinned npx clawhub execution command in the body of the skill, creating the same supply-chain execution risk. Users may copy-paste the command directly, causing arbitrary package code from an unversioned upstream release to execute in their environment if the package is ever hijacked or updated maliciously. The manual-skills coaching context is unrelated and therefore does not justify or mitigate the risk.

Content

No source excerpt is available for this finding.