Security audit

Creek

Security checks for vulnerabilities and agentic risk

Overview

This Creek deployment skill is coherent, but it tells agents to skip confirmations while managing live deployments, secrets, and domains.

Review before installing. Use this skill only for Creek projects, verify the npm package source/version, avoid running the CLI with elevated privileges, and require explicit confirmation before rollback, env value display/removal, domain changes, or any command that uses --yes.

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: SKILL.md:8
Finding: Unpinned Global Installation of a Third-Party CLI Package
Content: View full analysis
Remediation: View remediation
``` Commit the generated lockfile and use `npm ci` in automated environments. 4. Verify package provenance and registry integrity metadata before installation. Where supported, require npm provenance attestations and review the package contents and lifecycle scripts. 5. Avoid executing dependency lifecycle scripts unless they are required and audited. For initial inspection, use: ```bash npm install --ignore-scripts --save-dev --save-exact creek@ ``` 6. Run the CLI with least privilege in an isolated environment. Provide only the files and credentials needed for the deployment, and never install the package with administrator or root privileges. 7. Rotate `CREEK_TOKEN` and review deployment activity if an untrusted package release may already have been installed or executed. ]]>

SkillSpector

Vulnerability Patterns

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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger text is broad enough to activate on generic deployment or troubleshooting requests that are not specifically about Creek. This can cause the agent to select and apply Creek-specific operational guidance in the wrong context, increasing the chance of unsafe commands, credential use, or changes to unrelated infrastructure.
Content: No source excerpt is available for this finding.

Autonomous Decision Making

Medium

Category: Excessive Agency
Confidence: 96% confidence
Finding: The instruction to follow breadcrumbs and use --yes encourages the agent to chain recommended commands from tool output with minimal human oversight. If breadcrumbs include risky next steps, the agent may autonomously perform consequential actions without adequate review, which is especially dangerous in deployment and infrastructure-management contexts.
Content: Scanner excerpt · SKILL.md (reported line 25)May include surrounding context.
md
1. **Always use `--json`** for structured output. Auto-enabled in non-TTY / CI. 2. **Follow `breadcrumbs`** in JSON responses — they suggest the next command. 3. **Use `--yes`** to skip confirmation prompts (auto-enabled in non-TTY). 4. **Check `ok` field** — `true` = success, `false` = error with `error` and `message` fields. ## Command Reference

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: Instructing the agent to always use --yes suppresses safety confirmations for operations that may be destructive or irreversible, such as rollback, environment-variable changes, or domain modifications. Combined with autonomous tool use, this reduces a key safeguard and makes accidental harmful actions more likely.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The command list includes sensitive and destructive actions like rollback, environment variable disclosure/removal, and domain removal without any warnings, guardrails, or confirmation requirements. In an agent setting, presenting these as routine actions can normalize unsafe execution and increase the risk of data exposure or service disruption.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.