Ae1
High
- Category
- analysis-evasion
- Content
`scripts/caniemail.mjs` queries the caniemail.com dataset. Node 22+ and nothing
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Security audit
Email Compatibility
Security checks for vulnerabilities and agentic risk
Overview
This skill is a coherent HTML email compatibility checker with disclosed, purpose-aligned network refresh and local caching behavior.
Before installing, understand that normal use may contact caniemail.com and store a local dataset cache for faster or offline checks. Use --offline if you want to avoid network access, and only pass email template files you are comfortable having a local linting tool read.
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
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (10)
Ae1
High
- Category
- analysis-evasion
- Content
`scripts/caniemail.mjs` queries the caniemail.com dataset. Node 22+ and nothing
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Ae1
High
- Category
- analysis-evasion
- Content
`scripts/caniemail.mjs` queries the caniemail.com dataset. Node 22+ and nothing
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Ae1
High
- Category
- analysis-evasion
- Content
`scripts/caniemail.mjs` queries the caniemail.com dataset. Node 22+ and nothing
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Ae1
High
- Category
- analysis-evasion
- Content
`scripts/caniemail.mjs` queries the caniemail.com dataset. Node 22+ and nothing
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Ae1
High
- Category
- analysis-evasion
- Content
`scripts/caniemail.mjs` queries the caniemail.com dataset. Node 22+ and nothing
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Intent-Code Divergence
Medium
- Confidence
- 95% confidence
- Finding
- At L020-L022, the file says the script 'works offline against a bundled dataset snapshot' and emphasizes 'Node 22+ and nothing else, with no dependencies and no install step.' Later, L178-L181 describes network fetching, caching, and explicit live refresh behavior. That is not merely incomplete detail; it creates an actively misleading picture of the tool's runtime behavior and data sources.
Context-Inappropriate Capability
Medium
- Confidence
- 89% confidence
- Finding
- The skill's stated purpose is checking email client compatibility using caniemail-backed data, but the implementation accepts an arbitrary dataUrl and fetches from it. Supporting custom mirrors may be operationally useful, yet it expands the skill into a general remote JSON retrieval capability not justified by the manifest text.
Description-Behavior Mismatch
Medium
- Confidence
- 92% confidence
- Finding
- The manifest frames the skill as writing/verifying HTML email and reporting compatibility issues backed by caniemail data, but this core file also fetches data from caniemail.com or arbitrary mirrors and persists it in a shared disk cache. While using support data is expected, the manifest does not disclose ongoing network access and filesystem cache writes as part of the skill's behavior.
Intent-Code Divergence
Low
- Confidence
- 76% confidence
- Finding
- The manifest describes a rendering-compatibility skill backed by caniemail.com data, which is compatible with network use, but the main usage section at L020-L022 strongly frames the script as offline-only. The later data-freshness section introduces live fetching and caching. This documentation split creates intent ambiguity about whether users should expect network access during normal use.
Static analysis
No suspicious patterns detected.