ClawHub
Back to skill

Security audit

Judgment Loop

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent decision-support prompt with evaluation tooling; the main caution is unpinned npx installation examples, not hidden runtime behavior.

This skill is reasonable to install if you want an agent to apply a decision and evidence-checking framework to consequential uncertain choices. Prefer ClawHub install or verify and pin the skills CLI before using the npx examples, and avoid the global non-interactive install option unless you intentionally want global Codex availability.

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:76
Finding

Unpinned npm Package Execution in Installation Instructions

Content
View full analysis
Remediation
View remediation
add sirsws/judgment-loop ``` 2. Apply the same pinned version consistently in `README.md` and `README.zh-CN.md`. 3. Document the expected npm package publisher, registry, and reviewed version so users can verify package provenance before execution. 4. Where supported, publish and verify package integrity hashes or signed provenance attestations. 5. Prefer a lockfile-backed local development dependency for reproducible automated installation rather than resolving the latest package at runtime. 6. Avoid global and non-interactive installation options unless they are necessary. Explain their effects and provide a lower-impact local installation command as the default. 7. Recommend inspecting the resolved package metadata before first use, particularly in CI or privileged environments. 8. Do not instruct users to run the installation command with `sudo` or an administrator account. ]]>

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (13)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The description presents a user-facing reasoning/decision aid for handling consequential uncertainty and evidence updates. The code shown does not implement that behavior; instead, it tests a separate scoring script's correctness and dataset coverage. Its primary purpose is software QA around evaluation cases and observations, including manifest validation and duplicate detection. That is materially different from the declared purpose, so this is a clear description-behavior mismatch.

Content

No source excerpt is available for this finding.

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.

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.

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.

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.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This markdown file describes auto-invocation when 'consequences are obvious + conclusion is uncertain,' but the condition is qualitative and open to interpretation. Without explicit trigger phrases, thresholds, or clearer negative examples beyond a few broad exclusions, the activation scope may be inconsistently or unintentionally expanded.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The default prompt includes a broad natural-language trigger phrase ('Use $judgment-loop to examine this choice...') and the policy allows implicit invocation, which increases the chance the skill will activate in contexts the user did not explicitly intend. Because this skill is designed for consequential choices and judgment under uncertainty, unintended activation could steer conversations, reshape advice, or inject an unnecessary decision framework into unrelated tasks.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

This JSON manifest-like file includes example turns such as 'What am I missing?' tied to positive activation, but it does not document any explicit trigger constraints or negative examples at the trigger-definition level. Because SQP-1 applies to manifest files, these broad natural-language examples can make activation scope ambiguous and overlap with common user requests.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
79% confidence
Finding

The example 'Should I leave a stable job...' is mapped to full activation, but the file provides no explicit rule explaining why this common decision-advice phrasing should trigger while similar everyday comparison or advice requests should not. In a manifest-like evaluation file, that leaves the trigger scope under-specified.

Content

No source excerpt is available for this finding.

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · LICENSE (reported line 12)May include surrounding context.

text
permit persons to whom the Software is furnished to do so.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE

Vague Triggers

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The README says the skill supports implicit activation when 'meaningful consequences and uncertainty coexist,' which is a subjective condition rather than a concrete trigger list. Although it includes some exclusion examples, the activation boundary remains open to interpretation and could cause unintended invocation in ordinary discussions involving uncertainty.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The natural-language test case explicitly instructs translation 'into Chinese,' which imposes a specific language choice in the skill data without any accompanying opt-in or locale-selection framing. SQP-3 applies to all file types, including JSON content containing natural-language strings.

Content

No source excerpt is available for this finding.