Security audit

AI DevBlog Factory

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent, but it creates and shares Feishu documents by default with incomplete and partly inconsistent permission disclosure.

Review Feishu permissions before installing. Use draft mode or require an explicit approval step before publishing, limit the Feishu bot to the intended folder, and assume generated reports may be visible to the requester or other workspace users depending on Feishu settings.

A.I.G

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

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The README prominently promotes automatic publication of generated content to Feishu documents, but the warning section only mentions permissions and copyright, not the privacy and disclosure risks of sending aggregated external content and AI-generated summaries into a team document platform. This can cause unreviewed or sensitive material to be automatically shared more broadly than intended, especially in enterprise environments where document visibility, indexing, and notifications increase exposure.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The trigger phrases are broad enough to match ordinary requests for technical blogs or weekly reports, which can invoke a workflow that fetches external content and publishes output to Feishu. Over-broad activation increases the chance of accidental execution of side-effectful actions without the user clearly intending document creation or external publication.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The skill documentation does not clearly warn users, near the execution flow, that running the skill results in publication to Feishu and document creation. Because the workflow has an external side effect, insufficient disclosure can cause users to unintentionally export generated content into a shared collaboration platform.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The skill's security note says Feishu documents are creator-only by default, but the workflow explicitly sets grant_to_requester: true, which changes document visibility and can share generated content with another principal. This mismatch can mislead users about who can access the published report, causing unintended disclosure of internal summaries, links, or sensitive operational context.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This manifest file contains the primary user-facing description in Chinese only, with no indication that other languages are supported or that the user can choose their preferred locale. The policy for natural-language violations applies to all file types and disallows forcing a specific language without opt-in.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.