Back to skill

Security audit

batch-id

Security checks for vulnerabilities and agentic risk

Overview

This skill does what it claims, but it handles bulk government ID data in ways that need review before installation.

Install only if you are authorized to process the ID documents and are comfortable sending image URLs to Tencent Cloud OCR. Treat the returned text file as highly sensitive personal data. The publisher should add explicit consent and privacy disclosures, use unique secure temporary files or in-memory attachments, clean up outputs after delivery, and pin reviewed dependency versions.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
batch_id.py:48
Finding

Predictable Shared Temporary File Exposes Sensitive Identity Data

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
requirements.txt:1
Finding

Unbounded Dependency Version Allows Unreviewed Future SDK Releases

Content
View full analysis
=3.0.0 ``` ### Technical Analysis The dependency declaration accepts every Tencent Cloud Python SDK release from version `3.0.0` onward. Consequently, two installations performed at different times can resolve to different code without any project change or security review. This Skill supplies Tencent Cloud credentials to the SDK and uses it to transmit identity-card image URLs to the OCR service. A compromised, malicious, or otherwise unsafe future package release satisfying `>=3.0.0` would therefore execute in a credential-bearing process. The package name is consistent with the declared Tencent Cloud integration, and the reviewed project contains no evidence that the current upstream package is malicious. The issue is the absence of deterministic version and artifact integrity controls, rather than confirmed malicious behavior by the dependency. ### Attack Path 1. A future release of `tencentcloud-sdk-python` is compromised, malicious, or contains a newly introduced exploitable flaw. 2. The release remains compatible with the broad requirement `>=3.0.0`. 3. During deployment or rebuild, the package installer selects that release. 4. The unreviewed package is imported by `batch_id.py`. 5. Its code executes inside the Skill process and can access the process environment, Tencent credentials passed to the SDK, submitted image URLs, and OCR results. This exploitation path requires compromise or insecurity in an allowed upstream release, the package distribution channel, or dependency resolution infrastructure. ### Impact Assessment A malicious dependency would execute with the same privileges as the Skill process. Depending on runtime controls, this could permit: - Theft of the Tencent Cloud SecretId and SecretKey supplied to the SDK. - Unau ...[truncated 492 chars]
Remediation
View remediation
``` - Generate and retain a lock file so transitive dependencies are also deterministic. - Require cryptographic package hashes during installation, such as through a hash-locked requirements file and `pip install --require-hashes`. - Install packages only from an approved package index over authenticated TLS. - Use automated dependency scanning and review security advisories before upgrading. - Upgrade dependencies through controlled pull requests rather than resolving new versions automatically during deployment. - Run the Skill with minimal filesystem and network permissions. - Continue using a dedicated Tencent Cloud identity with only the OCR permissions required by this Skill, and rotate credentials if dependency compromise is suspected. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (7)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill processes highly sensitive government ID data and explicitly returns extracted name, sex, and ID number in a downloadable text file, but the description does not clearly warn users about the sensitivity and redistribution risk of that output. This increases the chance of accidental exposure, insecure sharing, or retention of plaintext identity data by users or downstream systems.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The documentation says the skill uses Tencent Cloud OCR and requires network access to the Tencent OCR endpoint, but it does not clearly warn users that uploaded ID images will be transmitted to a third-party cloud service for external processing. For identity documents, failing to disclose third-party processing creates serious privacy, consent, and compliance risks.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill sends user-supplied ID-card image URLs to Tencent Cloud OCR and extracts highly sensitive personal data including name, sex, and ID number. In a skill context, transmitting government ID data to a third-party processor without clear disclosure, consent, minimization, or retention controls creates a significant privacy and compliance risk, especially because users may not realize their data leaves the local environment.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The code writes extracted ID-card details to a plaintext file in /tmp and returns that file for downstream delivery, exposing sensitive identity data at rest and in transit within the platform workflow. Even though /tmp is the writable directory in SCF, storing full identity records in an attachment increases the chance of unintended disclosure, cross-process access, or retention beyond user expectations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly processes up to 50 ID-card images, extracts their contents, generates a .txt file, and sends that file back to the user, yet the description provides no warning that this involves highly sensitive personal data. Omitting this notice increases the risk of users unknowingly submitting regulated PII and of downstream insecure handling, storage, or transmission of bulk identity data.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

All user-facing instructions and examples in the skill file are presented only in Chinese, which can amount to forcing a specific language without opt-in. The stated policy requires flagging language or locale constraints unless the skill offers a choice or clearly documents a justified region-specific limitation.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
95% confidence
Finding

The dependency is specified with only a lower bound, which allows installation of any newer version, including future major releases that may introduce breaking changes or vulnerable transitive behavior. While this is a common packaging practice, it weakens build reproducibility and increases supply-chain risk because different environments may resolve to different versions over time.

Content

Scanner excerpt · requirements.txt (reported line 1)May include surrounding context.

text
tencentcloud-sdk-python>=3.0.0