Back to skill

Security audit

Botpress

Security checks for vulnerabilities and agentic risk

Overview

The Botpress operations are read-only, but setup can execute unverified downloads and introduce broader changes to installed agent skills.

Review the setup before enabling this skill. Install a reviewed, verified CLI release separately and prevent automatic installation during ordinary Botpress requests. Check whether that release modifies other agent skills; the publisher documents OO_SKILLS_SYNC_DISABLED as a control. Connecting Botpress also entrusts OOMOL with account access and handling requested data. The downloaded installers could not be inspected, so their full effects remain uncertain.

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
99% confidence
Finding

The supplied command runs an installer from an unrelated domain instead of searching or reading Botpress data. No connection to Botpress is established in the code, so this is an undeclared installation and execution capability, not a demonstrated supporting implementation detail. The downloaded script's contents and further effects are not visible.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

Authentication can support data access, but the supplied skill consists entirely of a login command and shows no Botpress-specific operation. Its implemented purpose therefore materially differs from the declared search/read capability. The command alone does not establish which service is being authenticated.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
98% confidence
Finding

The curl-to-Bash pipeline immediately executes whatever script the installation endpoint returns, with the invoking user's permissions and no pinned version or artifact verification. HTTPS protects transport but does not prevent a compromised publisher or distribution endpoint from delivering code that accesses local files, credentials, or persistent configuration; this is the same underlying issue as the setup-workflow finding.

Content

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

The setup workflow instructs execution of downloaded installation scripts without warning about the system-integrity risk.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

The setup instructions pipe remotely downloaded content directly into Bash or PowerShell without a review, integrity-verification, or explicit installation-approval step. Although limited to a missing-CLI condition, this can turn a read-only Botpress request into arbitrary local code execution if the distribution endpoint or installer is compromised.

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