Back to skill

Security audit

Fiverr Gig Automation

Security checks for vulnerabilities and agentic risk

Overview

The skill is openly about Fiverr automation, but it needs Review because it asks for reusable Fiverr credentials and can automatically act from the user's business account without clear safeguards.

Install only if you are comfortable granting automation access to your Fiverr account. Before use, require manual approval for outbound messages, offers, gig creation, and review requests; avoid storing the Fiverr password in ordinary settings unless the platform provides secret storage; enable MFA; and confirm the automation complies with Fiverr account and messaging policies.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:74
Finding
Insufficiently Protected Fiverr Credential Storage<![CDATA[ ## Vulnerability Details **File Location**: `SKILL.md`, lines 74–76 **Vulnerability Type**: Plaintext sensitive credential storage guidance **Risk Level**: Medium ### Vulnerable Code Snippet ```markdown 2. Save Fiverr login email as `FIVERR_EMAIL` in [Settings > Advanced](/?t=settings&s=advanced) 3. Save password as `FIVERR_PASSWORD` 4. Save your Fiverr username as `FIVERR_USERNAME` ``` ### Technical Analysis The setup instructions direct users to save their Fiverr password in application settings without requiring a dedicated secret manager, encryption at rest, access controls, log redaction, or credential lifecycle protections. A password is a reusable authentication secret and should not be handled as ordinary configuration data. If the referenced settings interface stores values in plaintext, exposes them to other skills or processes, includes them in diagnostics, or serializes them into backups, the Fiverr account credentials could be disclosed. The supplied project does not contain implementation code demonstrating secure storage or limiting access to `FIVERR_PASSWORD`. ### Attack Path 1. A user follows the documented setup procedure and stores their Fiverr email, username, and password in the application settings. 2. The settings backend, another installed component, a diagnostic export, a log, or a user with access to the settings obtains the stored value. 3. The exposed password is used to authenticate to Fiverr as the victim. 4. Subject to Fiverr's authentication controls, the attacker may access messages, customer information, gigs, orders, and account settings. This attack path is conditional on the settings platform exposing or inadequately protecting the stored value; the project does not provide the settings implementation needed to establish that behavior directly. ### Impact Assessment Successful credential disclosure could allow unauthorized access to the associated Fiverr seller account. The potential scope includes readi ...[truncated 445 chars]
Remediation
<![CDATA[ ## Remediation Suggestions 1. Store `FIVERR_PASSWORD` only in a dedicated secret-management facility or operating-system credential store, never in ordinary settings or committed configuration files. 2. Document whether the referenced settings interface encrypts secrets at rest and restricts secret retrieval to the specific skill that requires them. 3. Ensure secret values are masked in the user interface and excluded from logs, telemetry, crash reports, backups, command output, and diagnostic exports. 4. Apply least-privilege access controls so unrelated skills, plugins, users, and processes cannot read the credential. 5. Prefer short-lived session credentials or an officially supported delegated authorization mechanism instead of retaining a reusable account password, where Fiverr supports such an option. 6. Require multifactor authentication on the Fiverr account and notify users that browser-automation sessions must be protected against cookie and profile theft. 7. Provide credential rotation and revocation procedures, including immediate password rotation if settings, logs, or backups may have exposed the secret. 8. Add implementation and tests that verify secret redaction, encrypted storage, restricted retrieval, and safe deletion. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
Findings (3)

Ae1

High
Category
analysis-evasion
Content
- `SKILL.md` — this file
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill explicitly automates sending buyer offers, inbox replies, and review requests on a user's Fiverr account, but the description does not clearly warn that it will take autonomous actions on the user's behalf. That omission increases the risk of unintended account activity, spammy behavior, policy violations, or reputational harm because a user may enable the skill without understanding the scope of automated external actions.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill asks the user to store Fiverr login credentials, including email and password, without any warning about secret handling, storage sensitivity, or the risks of browser-based account automation. Because these credentials provide direct access to the user's Fiverr account, poor handling could lead to account takeover, unauthorized messaging, financial loss, and compromise of associated personal data.

Static analysis

No suspicious patterns detected.