Security audit

Dependency Audit

Security checks for vulnerabilities and agentic risk

Overview

This is a dependency-audit template with some risky unpinned tool examples, but no hidden execution, persistence, or deceptive behavior.

Before installing or using this skill, treat its commands as templates rather than safe defaults. Pin audit tools and GitHub Actions to reviewed versions or commit SHAs, run dependency-changing commands only after reviewing lockfile diffs, and get legal review for license decisions.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
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

Findings (2)

T08 · Insecure Dependencies

Warning

Location: SKILL.md:255
Finding: Unpinned Third-Party Security Tools and Mutable CI Action References
Content: View full analysis
Remediation: View remediation
``` ```bash python -m pip install --require-hashes -r audit-tools.lock go install golang.org/x/vuln/cmd/govulncheck@ npm ci npx --no-install license-checker ... ``` ]]>

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md:202
Finding: Forceful Automated Dependency Rewriting Can Introduce Unreviewed Breaking Changes
Content: View full analysis
Remediation: View remediation
@ --save-exact npm test git diff -- package.json package-lock.json ``` ]]>

SkillSpector

Vulnerability Patterns

Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (3)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The manifest description lists several broad activation phrases such as 'review package security' and 'check license compliance' without defining clearer trigger boundaries or exclusions. Because this is a manifest file section in a markdown file, the activation guidance is ambiguous enough to overlap with general security-review requests rather than a narrowly scoped dependency-audit intent.
Content: No source excerpt is available for this finding.

Rp1

Medium

Category: MCP Rug Pull
Confidence: 94% confidence
Finding: The skill recommends running npx license-checker without pinning a specific version, which allows resolution of the latest published package at execution time. If the upstream package or dependency chain is compromised, users of the skill could unknowingly execute attacker-controlled code during a security audit workflow.
Content: No source excerpt is available for this finding.

Autonomous Decision Making

Medium

Category: Excessive Agency
Confidence: 75% confidence
Finding: Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
Content: Scanner excerpt · SKILL.md (reported line 343)May include surrounding context.
md
- [ ] Do not report only direct dependencies — transitive dependency vulnerabilities are often more dangerous and are the most commonly missed - [ ] Do not present raw audit tool output without interpretation — a table of 200 CVEs with no prioritisation is worse than no audit at all - [ ] Do not assign all Critical CVEs as "fix immediately" without checking whether an exploitable path exists in your usage context - [ ] Do not make license compliance decisions without legal input — flagging a GPL dependency without a recommendation is incomplete work - [ ] Do not complete the audit without including a CI/CD pipeline step — a one-time audit that leaves the door open for new vulnerabilities is not a remediation

Static analysis

No suspicious patterns detected.