Security audit

letpub-skills

Security checks across malware telemetry and agentic risk

Overview

This skill does what it claims, but it asks users to store LetPub passwords and session cookies in plaintext and automatically reuses them without enough safety guidance.

Only install this if you are comfortable giving the skill a LetPub account login. Use a dedicated low-risk LetPub account if possible, do not reuse an important password, keep credentials.json and cookies.json out of source control and shared folders, and delete those files when you are done.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Context-Inappropriate Capability

Medium

Confidence: 89% confidence
Finding: The skill’s stated purpose is journal querying and recommendation, but this file adds account login, credential ingestion from disk, and persistent session-cookie handling. That expands the trust boundary to sensitive authentication material without clear necessity, increasing the risk of account misuse, credential exposure, or unauthorized authenticated scraping if the skill is run in shared or automated environments.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The instructions tell users to place raw account credentials in a local JSON file and allow the script to automatically transmit them for re-login, but there is no warning about secret storage risks, file protection, or network handling. Storing credentials this way increases exposure to local compromise, accidental inclusion in source control, and misuse by other tools that can read the workspace.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill instructs users to copy authenticated LetPub session cookies from their browser and store them in a local JSON file, but provides no warning that these cookies are effectively bearer credentials. Anyone who obtains the file can likely reuse the session to access the user's account, and the guidance normalizes insecure credential handling in plaintext.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The code reads credentials from a local JSON file and automatically transmits them to a remote service, but provides no explicit warning, consent flow, or secure-secret handling guidance. In the context of an agent skill, this can cause users to place real account secrets into project assets, increasing the chance of accidental disclosure through source control, backups, or multi-user environments.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The code serializes authenticated session cookies to disk in assets/cookies.json without encryption, access controls, or warning that the file contains reusable authentication material. If another local user, process, backup system, or accidental commit accesses this file, the session may be replayed to impersonate the account without knowing the password.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: scripts/login.py:151

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: SKILL.md:75