ClawHub
Back to skill

Security audit

Nexus

Security checks for vulnerabilities and agentic risk

Overview

Nexus is a repository-mapping and context-pack skill whose file reads and local artifact writes are disclosed and aligned with its purpose, with some install and data-handling cautions.

Before installing, prefer a pinned or otherwise verified installer path instead of a moving `npx` target. Use the skill only on repositories you are authorized to inspect, choose a private output directory, and review generated maps and context packs before sharing because names and selected code context can still reveal sensitive project details.

A.I.G

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
README.md:19
Finding

Unpinned Third-Party Installer Executed Through npx

Content
View full analysis
Remediation
View remediation
add AntreasAntoniou/nexus --skill nexus ``` 2. Document the expected npm package publisher, registry, version, and provenance so users can verify that the resolved package is authentic. 3. Provide an integrity-protected installation method, such as a release artifact with a published SHA-256 checksum or signature. 4. Prefer installation from an immutable, reviewed Git commit rather than a moving branch: ```sh git clone git checkout ``` 5. Recommend reviewing the resolved npm package before first execution, particularly in environments containing sensitive repositories or credentials. 6. In automated environments, use a lockfile or controlled internal registry and deny unexpected lifecycle scripts or package-version changes. 7. Run installation with the least-privileged account appropriate for the task and avoid elevated privileges. ]]>

SkillSpector

Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · tests/test_mapper.py (reported line 41)May include surrounding context.

python
(self.root / "a.py").write_text("def first(): pass\n")
        (self.root / "src").mkdir()
        (self.root / "src" / "ok.py").write_text("pass\n")
        (self.root / ".env").write_text("SYNTHETIC=hidden")
        (self.root / "secret_config.py").write_text("def private(): pass\n")
        (self.root / "node_modules").mkdir()
        (self.root / "linked").symlink_to(self.root, target_is_directory=True)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · tests/test_mapper.py (reported line 50)May include surrounding context.

python
(self.root / "a.py").write_text("def first(): pass\n")
        (self.root / "src").mkdir()
        (self.root / "src" / "ok.py").write_text("pass\n")
        (self.root / ".env").write_text("SYNTHETIC=hidden")
        (self.root / "secret_config.py").write_text("def private(): pass\n")
        (self.root / "node_modules").mkdir()
        (self.root / "linked").symlink_to(self.root, target_is_directory=True)

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill instructs the agent to read from the repository and write multiple artifacts (context-pack.md, directory-map.md, sources.txt), but it does not declare an explicit tool scope such as permissions or allowed-tools. That creates a capability/constraint mismatch: a host may grant broader file read/write access than intended, and the skill itself also directs reading repository instructions and writing outside the repository, which increases the chance of unauthorized data access or disclosure if the runtime does not enforce strict boundaries.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

This code performs a file write operation by creating and writing an output file, but there is no user-facing warning before the write occurs. The script does print a success message afterward, yet the rule asks for some form of disclosure around safety-relevant file modifications, and the write itself is not preceded by confirmation or an explicit warning in the interaction flow.

Content

No source excerpt is available for this finding.

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
tests/test_mapper.py:12