ClawHub
Back to skill

Security audit

version-updates

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed release-version workflow that can edit multiple repository files, but the behavior fits its stated purpose and shows no hidden exfiltration or persistence.

Install this only if you want agent assistance with release/version updates. Because the triggers are broad and the workflow may update many config and documentation files, review the dry-run output and git diff carefully before allowing changes to proceed.

A.I.G

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

Context Leakage

High
Category
Data Exfiltration
Confidence
75% confidence
Finding

Code or instructions that leak agent conversation context to external services, potentially exposing sensitive user interactions.

Content

Scanner excerpt · SKILL.md (reported line 39)May include surrounding context.

md
4. `version-update:docs-updated`
5. `version-update:verification`

## Step 1: Collect Context (`context-collected`)
- Confirm which version to apply (default: bump patch).
- If the prompt provides an explicit version, note it.
- validate `Skill(sanctum:git-workspace-review)` has already captured the repository status.

Direct Prompt Extraction

High
Category
System Prompt Leakage
Confidence
85% confidence
Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · SKILL.md (reported line 100)May include surrounding context.

md
- Run relevant builds or tests if version bumps require them (e.g., `cargo test`, `npm test`).
- Show `git status -sb` and `git diff` excerpts to confirm the version bumps.

## Output Instructions
- Summarize the files changed and the new version number.
- Mention follow-up steps, such as publishing or tagging, if applicable.

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
92% confidence
Finding

Using 'version' as a trigger risks colliding with a built-in command of the same name, causing ambiguous routing or unintended invocation of this skill instead of the expected command. Because this skill can update versions across many files and run scripts, a trigger collision can have meaningful operational impact.

Content

No source excerpt is available for this finding.

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
90% confidence
Finding

Using 'release' as a trigger similarly risks shadowing a built-in command and may cause this skill to activate during ordinary release-related interactions. In context, the skill is empowered to coordinate broad repository edits, so mistaken invocation can lead to unwanted version bumps, changelog edits, or follow-on workflow steps.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger list includes very broad single-word activators such as 'version', 'release', 'changelog', and 'bump', which are likely to match ordinary user conversation and invoke the skill unintentionally. In a skill that performs repository-wide version and documentation changes, accidental activation can lead to unwanted edits across many files and potentially disruptive release workflow actions.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The workflow directs the agent to perform automated bulk version and documentation updates, including use of a repository-wide script and recursive file discovery, without an explicit safety warning that many files may be modified. This increases the chance of surprising or excessive changes, especially in nested packages and documentation references.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.