Back to skill

Security audit

Contentful

Security checks for vulnerabilities and agentic risk

Overview

The Contentful workflow is controlled, but its setup runs unpinned remote software whose broader changes to agent skills are not explained.

Review the CLI installation separately before allowing setup, and verify the chosen release. Its current documentation provides OO_SKILLS_SYNC_DISABLED=1 to prevent startup writes to agent skill directories: [CLI controls](https://github.com/oomol-lab/oo-cli/blob/main/docs/commands.md#environment-variables). Use narrowly scoped Contentful access and retain the required write confirmations. The credential-theft and autonomous-billing warnings were unsupported, but the remote installer and broader persistent effects remain material concerns.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • 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
Findings (6)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied command executes an external installer and contains no Contentful operations or demonstrated connection to Contentful functionality. On the supplied evidence, this is an undeclared installation capability involving an inconsistent resource, rather than an established supporting implementation detail. The downloaded script's contents and further effects are not available for assessment.

Content

No source excerpt is available for this finding.

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 41)May include surrounding context.

md
## Available actions

- `create_entry` — Create a Contentful entry in a specific environment. [write]
- `get_current_user` — Get the authenticated Contentful user profile for the current personal access token.
- `get_entry` — Get a single Contentful entry by identifier.
- `list_content_types` — List Contentful content types available in a specific environment.
- `list_entries` — List Contentful entries with common filtering, pagination, and include options.

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 46)May include surrounding context.

md
- `list_content_types` — List Contentful content types available in a specific environment.
- `list_entries` — List Contentful entries with common filtering, pagination, and include options.
- `list_environments` — List Contentful environments inside a specific space.
- `list_spaces` — List Contentful spaces accessible to the current personal access token.
- `update_entry` — Update a Contentful entry using optimistic locking. [write]

## Safety

External Script Fetching

High
Category
Supply Chain
Confidence
99% confidence
Finding

Piping an HTTPS response directly into Bash executes mutable remote content without independently verifying its integrity or provenance. Compromise of the installer distribution infrastructure could therefore compromise the agent's host, although the provider-aligned URL and conditional setup context do not suggest deliberate malicious intent.

Content

Scanner excerpt · SKILL.md (reported line 62)May include surrounding context.

The setup workflow executes remotely downloaded installation scripts without warning about the system-integrity implications.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

The setup instructions direct the agent to execute downloaded shell or PowerShell code without explaining the host-level consequences or requiring installation-specific approval. A compromised installer could execute with the invoking user's privileges, accessing local files and credentials; restricting installation to a missing-CLI error reduces exposure frequency but not impact.

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 83)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