Back to skill

Security audit

GL importer

Security checks for vulnerabilities and agentic risk

Overview

This skill is a legitimate accounting importer, but it exposes live accounting changes, company settings updates, mapping deletion, and import rollback without strong in-skill confirmation boundaries.

Install only if you are comfortable letting an agent operate against live accounting data through Synder. Use a limited/revocable API token, start with dryRun=true, review mappings and medium-confidence matches, and require explicit approval before dryRun=false imports, settings changes, mapping deletion, cancellation, or reversion.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The revert endpoint is a destructive capability that can delete entities previously created in the accounting system. Because the skill is positioned as an importer, exposing deletion behavior creates a high-risk mismatch where an agent or prompt injection could trigger irreversible accounting changes under the guise of import assistance.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

The mapping deletion endpoint introduces a mutable administrative action that can be abused through overbroad tool access or parameter manipulation. Even as a soft delete, disabling mappings can disrupt established import workflows, cause operational failures, or force remapping in ways the user did not request.

Content

Scanner excerpt · references/api.md (reported line 167)May include surrounding context.

md
### PUT /companies/{id}/mappings/{mid}
Same body as POST. Returns 200.

### DELETE /companies/{id}/mappings/{mid}
Soft delete (sets active=false). Returns 204.

### GET/POST /companies/{id}/settings

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 67)May include surrounding context.

1. Find the company

bash
curl "$BASE/companies" -H "Authorization: Bearer $IMPORTER_API_TOKEN"

Response is an array. Pick the company with status: "ACTIVE". Save its id.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The API reference exposes capabilities that materially exceed the skill's declared import-only purpose, including account/company data retrieval, settings modification, mapping deletion, and import reversion. In an agent setting, this scope mismatch increases the chance that the skill can be invoked for unintended actions or data access beyond what a user expects from an importer.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The settings update endpoint allows modifying company-level import and accounting behavior, which is outside the narrow core workflow of uploading and processing a file. Misuse could silently alter future import semantics or accounting defaults, leading to integrity issues that are difficult to detect and recover from.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.