Back to skill

Security audit

Einstein Research Suite

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed market-research skill pack, but it should be reviewed because it tells users to install 11 mutable external skills that are not pinned or bundled for inspection.

Before installing, review each referenced skill separately and prefer pinned versions or verified publisher/source information. This package itself did not show malicious behavior, but following its quick start changes the agent by adding 11 external skills whose exact contents are not locked by this artifact.

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:39
Finding

Unpinned Third-Party Skill Installations Create a Supply-Chain Risk

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 39–51
Vulnerability Type: Unpinned third-party skill dependencies
Risk Level: Medium

The package directs users to install 11 externally hosted ClawHub skills without immutable version constraints, content digests, signatures, or canonical source references.

bash
clawhub install einstein-research-backtest-dv
clawhub install einstein-research-backtest-engine-dv
clawhub install einstein-research-breadth-dv
clawhub install einstein-research-bubble-dv
clawhub install einstein-research-edge-dv
clawhub install einstein-research-ftd-dv
clawhub install einstein-research-options-dv
clawhub install einstein-research-portfolio-risk-dv
clawhub install einstein-research-regime-dv
clawhub install einstein-research-scenario-dv
clawhub install einstein-research-themes-dv

Technical Analysis

These commands resolve mutable package names through an external registry. The repository does not pin versions or content hashes and does not include the referenced skill implementations for local review. Consequently, the content installed by a user may differ over time from the content originally intended by the package author.

This is a supply-chain weakness rather than evidence that any referenced package is currently malicious. Exploitation would require compromise or malicious control of a referenced package, its publisher account, the registry, or the dependency-resolution process. Without integrity verification, users cannot reliably establish that downloaded skill content matches a previously audited artifact.

Attack Path

  1. An attacker compromises a referenced skill publisher account or the ClawHub distribution channel, or otherwise gains control over the package resolved by one of the listed names.
  2. The attacker publishes a modified skill under that package name.
  3. A user follows the documented quick-start procedure without specifying an imm ...[truncated 1031 chars]
Remediation
View remediation

Remediation Suggestions

  1. Pin every referenced skill to an immutable release version rather than resolving only by package name.
  2. Prefer content-addressed dependencies and verify cryptographic hashes before installation.
  3. Require signed releases and validate signatures against documented publisher identities.
  4. Publish canonical source repository and release links for each included skill.
  5. Audit and vendor the referenced skill contents when feasible so the reviewed implementation is distributed with the pack.
  6. Maintain a lock file recording exact versions, digests, and resolved sources.
  7. Add automated integrity and provenance checks to the installation workflow.
  8. Apply least-privilege permissions and sandboxing to all installed skills, especially those capable of filesystem, network, shell, or credential access.
  9. Document that users must review dependency changes before upgrading to newer releases.
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