Back to skill

Security audit

ByteForms

Security checks for vulnerabilities and agentic risk

Overview

The ByteForms reads fit the stated purpose, but setup runs an unchecked remote installer and the required CLI can silently change other agent skills.

Review automatic setup before enabling this skill. Use a separately reviewed, pinned CLI installation and set OO_SKILLS_SYNC_DISABLED=1 before invoking it if you do not want other agent skills installed or refreshed. Connect only the intended ByteForms account and consider whether its form responses may pass through OOMOL. The remote installer and exact runtime release were not verified.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The supplied chunk performs remote script execution rather than the declared ByteForms data operations. Although CLI installation could support an integration, the supplied code establishes no connection between this installer and ByteForms. The downloaded script's further behavior is not visible and cannot be assessed.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
86% confidence
Finding

The supplied chunk only initiates authentication; it contains no ByteForms-specific operation or data access. Authentication can be a supporting implementation detail, but this chunk provides no evidence that the login supports ByteForms, and its sole demonstrated purpose differs from the declared functionality.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
97% confidence
Finding

Piping a live remote installer directly into Bash executes its contents with the agent process's privileges without a pinned version or independent integrity verification; the PowerShell example has the same weakness. Compromise of the distribution endpoint could therefore enable local data theft or modification, although the documented vendor domain and missing-CLI condition provide no evidence of deliberate malicious intent.

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

The setup workflow executes downloaded installer scripts without warning about local system changes.

Medium
Category
Output Handling
Confidence
92% confidence
Finding

The setup instructions direct installation after a missing-command error without explaining local changes or requiring explicit approval. Installing software executes code and can alter the user's environment, exceeding the expected scope of a read-only ByteForms task; the conditional setup trigger limits exposure but does not establish consent.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% 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 78)May include surrounding context.

md
- **`OOMOL_INSUFFICIENT_CREDIT`** — OOMOL billing stop. Recharge at `https://console.oomol.com/billing/token-recharge` before retrying.

- **HTTP 402 without `OOMOL_INSUFFICIENT_CREDIT`** — the third-party account itself has no active plan or credit. Report the provider error message as is; do not ask the user to recharge OOMOL.

## Resources