This report does not block or authorize the manifest or ZIP.
AI review found most of the 555 static alerts are documentation or template false positives, especially markdown backticks, WordPress helper names, and example URLs. Two generated PHP templates contain moderate copy-paste risks: REST meta values are saved without explicit sanitization, and taxonomy custom field saves lack nonce verification. No prompt injection or malicious intent was found, so the skill is not blocked but should be published with a security warning.
Report position
Historical report
Open audit history before using this report to install.
Audit attestation
Not attestable
The required immutable binding is incomplete.
Human verification
Not verified
No human verification is recorded for this report.
Coverage
15 Files scanned · 5,786 Lines analyzed
3 items shown for review
Limitations
This report does not claim runtime or sandbox execution and does not prove the absence of side effects.
01
Evidence chain
Follow the evidence from source binding to the install contract. Available evidence supports verification; it is not a safety guarantee.
REST Template Saves Custom Meta Without Explicit Sanitization
The generated REST endpoint template writes price, location, and featured request parameters directly to post meta in create and update paths. Permission callbacks exist, but these meta parameters are not declared with sanitize callbacks before persistence.
The write paths are direct and semantic context confirms they persist request-controlled values. Legitimate use is possible after adding field-specific validation and sanitization.
Taxonomy Custom Field Save Lacks Nonce Verification
The taxonomy template saves custom term fields from POST data with sanitization, but the save handler does not verify a nonce before updating term meta. WordPress capability checks help, but nonce verification is still expected for admin form intent.
The handler clearly reads POST data and updates term meta without a visible nonce check in the template. Impact is moderate because values are sanitized and WordPress admin permissions still apply.
Scaffold Script Writes Files to User-Selected Paths
The scaffold script creates the requested output directory and writes generated PHP files there. This is expected functionality, but users should avoid running it with untrusted arguments or privileged output paths.
The filesystem behavior is confirmed in the script and is legitimate for a code generator. Risk depends on how the user invokes the script.
04
Expert evidence
Immutable subject identity, scanner metadata, dismissed matches, and source-level evidence.
Artifact subject
Marketplace commit
Unavailable
Content hash
Unavailable
Tree hash
Unavailable
Skill path
Unavailable
Audit payload hash
Unavailable
Analysis metadata
Audit model: codex
Analysis state: Complete
Scope is limited to the recorded files, lines, methods, and evidence. No runtime or sandbox execution is claimed.
Static Command Execution Alerts Are Documentation Examples
The many Ruby backtick and command substitution alerts are caused by markdown fences, inline WordPress examples, and shell commands shown as documentation. I did not find runtime shell execution in the scaffold script or templates.
Reviewed examples are fenced documentation, and the executable Python script does not invoke subprocess or shell APIs. This strongly supports a false-positive verdict for the command-execution cluster.
Static Weak-Crypto Alerts Are Mostly WordPress Terminology False Positives
The weak cryptography alerts align with WordPress terms, placeholders, markdown text, and helper names rather than confirmed MD5, SHA1, or custom crypto usage. No evidence found of cryptographic routines used for security decisions.
The cited areas contain metadata, constants, or security guidance, not weak hashing logic. Confidence is high, though this summarizes a large static cluster rather than every individual line.
Network Alerts Are External References and WordPress API Examples
Hardcoded URLs, fetch examples, email references, and REST examples are documentation or local WordPress integration patterns. No evidence found of credential exfiltration or unauthorized outbound network calls.
The reviewed network indicators are examples or official documentation links. The skill does not contain code that sends secrets to an external endpoint.