Versioned security assessment

Report ID: SA-33B07094

4/14/2026, 8:45:04 AM

pexoai-agent security assessment v22

Skill Security Certification Report

Audit History

Audit model: claude Historical report

Skill name: pexoai-agent
Version: v22
Maintainer: pexoai
Source: github.com/pexoai/pexo-skills/tree/main/skills/pexo-agent/
Coverage: 13 Files scanned · 2,110 Lines analyzed
Policy version: Unavailable

Highest confirmed finding severity

High

1 confirmed security finding requires attention.

Installation context

Historical evidence

This report may not describe the currently installable artifact. Open the current Skill page for install guidance.

Open current Skill page

This report does not block or authorize the manifest or ZIP.

This skill is an API client for the Pexo AI video platform. Static analysis detected 534 potential security patterns across 13 files, but the majority are false positives reflecting standard CLI tool behavior. The skill uses shell scripts to make authenticated HTTP requests, read local configuration, and manage temporary files. Detected patterns (external commands, network calls, filesystem access, environment variable reads) are all expected for an API client skill. Weak cryptographic algorithms (MD5/SHA1) are used for file checksums, which is low-risk in this context. No credential exfiltration, unauthorized network access, or malicious patterns were found after manual evaluation.

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

13 Files scanned · 2,110 Lines analyzed

6 items shown for review

Limitations

This report does not claim runtime or sandbox execution and does not prove the absence of side effects.

Evidence chain

Follow the evidence from source binding to the install contract. Available evidence supports verification; it is not a safety guarantee.

Source

Binding unavailable
Artifact

Identity incomplete
Audit

Complete
Install contract

Open manifest to verify
Open manifest

Capabilities observed

Observed means this report recorded supporting evidence. Not recorded does not prove that a capability is absent.

Contains scripts

May execute code included with the Skill.

Not recorded by this audit

Network access

May connect to external services.

Observed in 2 evidence locations

Filesystem access

May read or write local files.

Observed in 2 evidence locations

Env variables

May read values from the process environment.

Observed in 2 evidence locations

External commands

May invoke commands or programs outside the Skill.

Observed in 4 evidence locations

Capability review items (5)

High

scripts/_common.sh:21-33

API Key Access from Local Configuration

The skill reads PEXO_API_KEY from a local config file (~/.pexo/config) and uses it in HTTP Authorization headers. This is expected behavior for an API client skill, but grants the skill access to the user's API credentials. If the skill were modified maliciously, it could exfiltrate the key.

Pattern confirmed: PEXO_API_KEY is sourced from config and used in Authorization header. This is legitimate API client behavior, but flagged as high-risk because the skill has access to credentials.

Medium

scripts/_common.sh:223-414 scripts/pexo-chat.sh:48-133 scripts/pexo-upload.sh:33-89 SKILL.md:30-314

External Shell Command Execution

The skill uses extensive shell command substitution ($(command)) and curl invocations to interact with the Pexo API. This is legitimate CLI tooling for an API client, but the pattern triggers static analysis due to the general risk of command injection if user input were passed unsafely to shell commands.

Pattern confirmed: shell commands are used for legitimate API calls and file operations. No user-input injection vectors found in the script source, but SKILL.md instructs the AI agent to execute commands which requires trust in the AI's behavior.

Medium

scripts/_common.sh:279 SKILL.md:9-10

Network Access to External API

The skill makes HTTP requests to pexo.ai and related URLs. This is expected for an API client, but the network access could be misused if the skill were modified to send data to unauthorized endpoints.

Pattern confirmed: curl commands target pexo.ai API endpoints. All URLs point to the official Pexo service, which is consistent with the skill's stated purpose.

Medium

scripts/_common.sh:7 scripts/_common.sh:37

Hidden File Access in Home Directory

The skill reads from ~/.pexo/config and writes to ~/.pexo/tmp. This is standard configuration and temporary file storage for a CLI tool, but access to hidden files in the home directory could be a vector for reading sensitive data if misused.

Pattern confirmed: config sourced from ~/.pexo/config and temp directory at ~/.pexo/tmp. These are the skill's own files, not unrelated hidden files.

Low

scripts/pexo-doctor.sh:157-162 references/TROUBLESHOOTING.md:36-71

System Information Commands in Diagnostic Scripts

The doctor and troubleshooting scripts use commands like uname, df, and free to diagnose system configuration issues. These are legitimate diagnostic tools for a setup wizard.

Pattern confirmed: system diagnostic commands used in doctor/troubleshooting scripts for legitimate setup assistance purposes.

Risk findings

Confirmed security concerns are separated from items that still need review.

Confirmed security concerns (1)

RISK-001 High

scripts/pexo-asset-get.sh:10 scripts/pexo-chat.sh:10 scripts/pexo-doctor.sh:17 scripts/pexo-upload.sh:10

Weak Cryptographic Algorithm Usage

Multiple scripts use MD5 or SHA1 for file checksums and hashing. These algorithms are cryptographically weak and should not be used for security-critical operations. In this context, they are used for file integrity checks and deduplication, not security.

Pattern confirmed: md5sum/sha1sum commands used in scripts for file hashing. Not security-critical in this context, but flagged as weak crypto.

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: claude

Analysis state: Complete

Scope is limited to the recorded files, lines, methods, and evidence. No runtime or sandbox execution is claimed.

Static false positives ignored (1)

Low

SKILL.md:9-243 references/SETUP-CHECKLIST.md:12-79

Hardcoded URLs in Documentation

The SKILL.md and reference files contain hardcoded URLs to pexo.ai documentation and API endpoints. These are legitimate service URLs, not malicious endpoints.

Pattern confirmed: all URLs point to pexo.ai or pexo-skills GitHub repository. These are legitimate documentation and API references.

Verify and export

JSON report CITATION.cff SARIF 2.1.0 Signed installation manifest

The manifest and lockfile bind install artifacts to cryptographic hashes. This integrity claim is separate from the security assessment.

Audit attestation: not_attestable