Back to skill

Security audit

JavDB CLI

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: the reported update-related flag is considered expected for its purpose, though Skillspector’s review was partial.

The overall verdict is benign, not a guarantee of safety. Consider that Skillspector’s analysis was partial because three local references could not be resolved, even though all eight listed components were scanned. Review the update guidance and install only if its features fit your needs.

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
Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · SKILL.md (reported line 89)May include surrounding context.

md
CLI `host=auto` verifies and reuses `route.json`; when invalid, it discovers startup candidates and selects a route. Fixed `mirror`, `main`, or absolute URLs bypass discovery. Do not confuse the CLI default with the SDK constructor's mirror default or persist an unreviewed URL.

Actual update resolves its own Release transport, ignoring data host selectors. It detects Homebrew, `go install`, or Release-archive installation; development builds refuse self-update. Use `--prerelease` only on request, and not for Homebrew. See [install.md](references/install.md) for verification boundaries and [troubleshooting.md](references/troubleshooting.md) for failures.

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · references/install.md (reported line 13)May include surrounding context.

md
For an existing installation, `javdb update --check --json` is the read-only inspection route. Actual `javdb update` uses detected Homebrew, `go install`, or the Release archive. Release replacement validates the signed manifest, source/version/platform, archive checksum, and extracted binary hash; do not bypass those checks or execute a downloaded candidate manually to replace verification.

Use `--prerelease` only for an explicit prerelease request; Homebrew updates do not support it. Development builds refuse self-update. Report failures without switching installation channel or replacing an existing executable by hand.