Back to skill

Security audit

Clawdhub

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent ClawdHub CLI helper, but it includes unattended forced bulk updates of installed skills without enough scoping or warning.

Review before installing if you are not comfortable with a skill that can install, update, and publish other skills. Prefer explicit skill names and pinned versions, avoid unattended `--all --no-input --force` updates unless you have reviewed the source registry and backup plan, and install the npm package as a non-administrative user from a trusted registry.

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:4
Finding
Unpinned Global npm Dependency Installation## Vulnerability Details **File Location**: `SKILL.md:4` and `SKILL.md:9-12` **Vulnerability Type**: Unpinned third-party dependency and unsafe global installation **Risk Level**: Medium ### Vulnerable Code ```yaml metadata: {"clawdbot":{"requires":{"bins":["clawdhub"]},"install":[{"id":"node","kind":"node","package":"clawdhub","bins":["clawdhub"],"label":"Install ClawdHub CLI (npm)"}]}} ``` ```bash Install ```bash npm i -g clawdhub ``` ``` ### Technical Analysis The Skill declares and recommends installation of the `clawdhub` npm package without an exact version or integrity constraint. Consequently, npm resolves whichever package release is current when installation occurs rather than the specific release reviewed during this audit. npm installation can execute package lifecycle scripts, including `preinstall`, `install`, and `postinstall`. Because the command uses global installation, those scripts run with the permissions of the invoking user and can modify the applicable global npm installation area. The issue does not prove that the current package is malicious; it creates a mutable supply-chain execution path through which a compromised or malicious future package release could execute code without changes to this Skill. ### Attack Path 1. An attacker compromises the npm publisher account, package distribution process, or another component capable of modifying a future `clawdhub` release. 2. The attacker publishes a release containing malicious package content or lifecycle scripts. 3. A user follows the documented `npm i -g clawdhub` instruction after the malicious release becomes the version selected by npm. 4. npm downloads the attacker-controlled release and executes any enabled lifecycle scripts. 5. The scripts execute with the invoking user's privileges and can access files, credentials, environment variables, and network resources available to that user. 6. If installation is performed from an el ...[truncated 841 chars]
Remediation
## Remediation Suggestions 1. Pin the dependency to an exact, reviewed version in both metadata and installation documentation, for example: ```bash npm install --global clawdhub@<reviewed-exact-version> ``` 2. Re-audit the package before changing the pinned version, including its bundled files, transitive dependencies, and lifecycle scripts. 3. Verify package provenance and integrity through trusted registry metadata, checksums, signatures, or npm provenance attestations where available. 4. Prefer a project-local installation with a committed lockfile instead of global installation when operationally feasible. 5. Install with lifecycle scripts disabled when the package does not require them: ```bash npm install --global --ignore-scripts clawdhub@<reviewed-exact-version> ``` 6. Perform installation as a non-administrative user and avoid `sudo` or privileged shells. 7. Restrict registry overrides such as `CLAWDHUB_REGISTRY` to explicitly trusted endpoints and document the approved registry.
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
Findings (1)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly documents `clawdhub update --all --no-input --force` without warning that it can non-interactively modify many local skill files at once. In an agent setting, this increases the risk of unintended mass changes, silent upgrades to unreviewed skill versions, and propagation of unsafe or incompatible code into the local environment.

Static analysis

No suspicious patterns detected.