📦

Audit History

xhs-creator-collaboration-evaluation - 3 audits

Version comparison

Capability and finding changes across audited versions, newest first.

VersionDateResultReview itemsChange vs previous
v3 LatestNov 2, 2026, 12:09 PM No confirmed findings2No capability change
v2 Oct 29, 2026, 04:56 AM No confirmed findings2No capability change
v1 Oct 21, 2026, 08:18 AM 2 confirmed0Baseline

Nov 2, 2026, 12:09 PM

Two command examples automatically execute an unpinned npm package, creating a supply-chain risk in an environment configured with an API key. The remaining 38 findings concern Markdown formatting, ordinary Chinese text, creator identifiers, service links, or documented authentication. No evidence found of prompt injection, credential exfiltration, or destructive intent in the supplied files; external package behavior was not reviewed.

2
Files scanned
141
Lines analyzed
5
Review items
0
False positives ignored
Capability review items (2)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Medium
Ruby/shell backtick execution · 2 occurrences
```bash
This fenced example runs npx -y socialdatax-skills@latest, automatically downloading and executing a mutable third-party package. A compromised release could access local files and the configured API key.
Audited by: codex

Oct 29, 2026, 04:56 AM

Of 40 static findings, 38 are false positives involving Markdown, Chinese text, official links, credential configuration, or creator identifiers. Two command examples execute an unpinned npm package with automatic installation, creating dependency risk in an environment containing the API key. No evidence found of prompt injection, credential exfiltration, or concealed payloads in the two reviewed files; external package implementation was not reviewed.

2
Files scanned
141
Lines analyzed
5
Review items
0
False positives ignored
Capability review items (2)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Medium
Ruby/shell backtick execution · 2 occurrences
```bash
The fence itself is Markdown, but its commands run npx -y socialdatax-skills@latest. This automatically executes mutable registry code with access to the configured API key.
Audited by: codex

Oct 21, 2026, 08:18 AM

All 39 static matches are false positives involving Markdown, legitimate authentication documentation, creator identifiers, and readable Chinese text. Separate semantic risks concern automatic execution of an unpinned npm package and forwarding payment links without destination validation. No evidence found of embedded malware, credential exfiltration, or prompt injection in the two supplied files; external package behavior was not reviewed.

2
Files scanned
141
Lines analyzed
5
Review items
0
False positives ignored

Confirmed security concerns (2)

Medium
Unpinned Third-Party Package Execution
CLI examples use 'npx -y socialdatax-skills@latest', automatically accepting a mutable npm release. A compromised release could execute locally with access to configured credentials.
The examples explicitly select latest and suppress the installation prompt with -y. This establishes a supply-chain exposure, not evidence that the package is malicious.
Medium
Unvalidated Payment Links From Error Responses
The instructions require displaying returned recharge links unchanged and encouraging payment. Without destination validation, a tampered response could redirect users to a phishing site.
The payment workflow explicitly trusts error-supplied URLs without a hostname check. The provider-domain restriction addresses API key management, not these returned billing links.
Audited by: codex