Back to skill

Security audit

Pinterest Browser Publisher

Security checks for vulnerabilities and agentic risk

Overview

The skill largely matches its Pinterest publishing purpose, but it handles login cookies insecurely and can publish content automatically under a real account.

Use this only with a Pinterest account you are comfortable automating. Treat ~/.config/pinterest/cookies.json as a login secret, do not share logs from force-login.js, delete or rotate cookies after use, review pins before running auto-publish scripts, and prefer pinned/local dependency installation in an isolated user environment.

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 (3)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/force-login.js:64
Finding

Insecure Storage and Logging of Pinterest Session Cookies

Content
View full analysis
c.name.includes('sess') || c.name.includes('token') || c.name.includes('auth') ); importantCookies.forEach(c => { console.log(` - ${c.name}: ${c.value.slice(0, 50)}...`); }); ``` Another affected login flow uses the same unprotected write pattern: ```javascript // Save cookies const cookies = await context.cookies(); fs.writeFileSync(COOKIES_FILE, JSON.stringify(cookies, null, 2)); console.log(`✓ Cookies saved to ${COOKIES_FILE}`); ``` ### Technical Analysis Pinterest authentication cookies are bearer credentials. Anyone who obtains a valid session cookie may be able to impersonate the authenticated user without knowing the account password or completing multi-factor authentication. The scripts serialize the complete browser cookie set to `~/.config/pinterest/cookies.json` without specifying an explicit file mode. Consequently, access permissions depend on the runtime environment's `umask`. On a permissively configured or shared system, other local users or processes may be able to read the file. In addition, `force-login.js` deliberately prints the first 50 characters of cookies whose names contain `sess`, `token`, or `auth`. These values can enter terminal history, CI output, remote support logs, shell-session record ...[truncated 1785 chars]
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
package-lock.json:13
Finding

Mutable and Non-Official Dependency Installation Sources

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Note
Location
scripts/auto-publish-all.js:176
Finding

Predictable Temporary Screenshots and Hard-Coded User Paths

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (53)

Tp4

High
Category
MCP Tool Poisoning
Confidence
89% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
89% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
86% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
89% confidence
Finding

The write-up claims support for jp.pinterest.com, batch publishing, carousel handling, and cookie persistence, while the analyzed behavior suggests a narrower single-pin workflow on a different host and only partial cookie management. In a skill that automates actions on a logged-in account, these discrepancies are security-significant because they obscure actual scope and side effects.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 42)May include surrounding context.

md
node scripts/publish-fix.js

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 56)May include surrounding context.

md
node scripts/publish-fix.js

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 79)May include surrounding context.

md
node scripts/publish-fix.js

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 45)May include surrounding context.

md
node scripts/auto-publish-all.js

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

md
node scripts/auto-publish-all.js

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The script logs partial values of cookies whose names suggest session, token, or authentication data. Even truncated secrets materially increase credential exposure risk through terminal history, CI logs, screen recording, shell capture, or shared support sessions, and this capability is not required to publish Pinterest content. In this skill context, persisted browser cookies are already sensitive; additionally printing them makes account takeover more likely if logs are accessed.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The installation command uses npx clawhub@latest, which executes whatever the latest published package version is at install time rather than a reviewed, fixed version. This creates a supply-chain risk: a compromised or maliciously updated package could run arbitrary code on the user's system during installation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill prominently advertises persistent Pinterest cookie storage and specifies a disk path for authenticated session data, but does not warn that stolen session cookies may allow account takeover without re-entering credentials. Because this is a browser automation skill that encourages long-term reuse of login state, the context makes the omission more dangerous, especially on shared or compromised machines.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
84% confidence
Finding

The skill documents persistent local storage of sensitive session material and credentials under ~/.config/pinterest/, including cookies and email/password. Session persistence is useful operationally, but if those files are weakly protected, copied, or committed by mistake, an attacker could reuse the session or recover credentials to access the Pinterest account.

Content

Scanner excerpt · README.md (reported line 68)May include surrounding context.

Configuration

Create ~/.config/pinterest/config.json:

json
{

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The README explicitly suggests using residential proxies to get past CAPTCHA and account checks, which is an evasion tactic rather than a requirement for ordinary pin publishing. In the context of browser automation against a consumer platform, this guidance facilitates bypassing anti-abuse controls and increases the likelihood of account misuse, policy violations, or scaled deceptive activity.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding

The skill describes browser automation that reads persisted cookies from the user's home directory, accesses local image files, and writes screenshots, but it declares no explicit tool scope or permissions. Missing scope declarations make the skill's actual capabilities less transparent to users and reviewers, increasing the risk of unintended local data access and misuse of session material.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The description states the skill supports jp.pinterest.com with Japanese copywriting, which imposes a language/locale behavior in natural-language instructions. The file does not clearly present this as an optional user-selected mode or explain a compliance-driven reason that would justify requiring Japanese output.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill instructs users to persist Pinterest login cookies in a predictable path under the home directory without warning that these cookies are effectively bearer tokens for the account. If another process, user, or tool can read that file, an attacker may hijack the Pinterest session and post or access account data without credentials.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The script performs the final publish action automatically using persisted Pinterest cookies and immediately clicks a detected publish/save button without requiring any runtime confirmation from the user. In this skill context, that means content can be posted to a real account irreversibly or at least with account-level consequences, making accidental or unauthorized publication materially risky.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script saves live Pinterest authentication cookies to disk in a predictable location under the user's home directory and then prints portions of sensitive values without a clear warning that these are credentials equivalent to a login session. A local attacker, malware, backup leak, or accidental file sharing could reuse these cookies to hijack the Pinterest account. Because this skill is specifically for browser-based publishing without an API key, cookie persistence is operationally useful, but that makes secure handling more important, not less.

Content

No source excerpt is available for this finding.