Pub Pdf

ReviewAudited by ClawScan on May 10, 2026.

Overview

This PDF-branded skill is really a broad SkillBoss API gateway that can use one API key for many providers, including email/SMS batch actions, so it needs review before installing.

Treat this as a broad SkillBoss AI gateway, not just a PDF utility. Install it only if you need those capabilities, use a dedicated limited API key, require explicit approval before any email/SMS or batch action, monitor spending, and avoid sending sensitive documents or media unless you accept the external provider data handling.

Findings (5)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

A user looking for a PDF utility may grant access to a much broader AI service than expected.

Why it was flagged

The PDF-focused name and first claim are paired with a much broader multi-model service. The provided artifacts do not show the claimed PDF merge/split/forms toolkit, while they do show broad non-PDF capabilities.

Skill content
name: pdf
description: "Comprehensive PDF manipulation toolkit for extracting text, creating, merging, splitting documents, and handling forms. And also 50+ models for image generation, video generation, text-to-speech, speech-to-text, music, chat, web search, document parsing, email, and SMS."
Recommendation

Install only if you intend to use the broader SkillBoss API gateway, not just a narrow PDF tool. The description should be narrowed or the non-PDF powers should be clearly highlighted before use.

What this means

If the API key permits these models, an agent could send unwanted messages or incur costs if prompted incorrectly or if it over-applies the skill.

Why it was flagged

The model catalog includes actions that can send emails and SMS messages, including batch sends. The provided artifacts do not define clear user-confirmation, recipient-scoping, rate-limit, or rollback requirements for these high-impact actions.

Skill content
`email/send` | Send single email |
| `email/batch` | Send batch emails |
...
| `prelude/notify-send` | Send SMS notification |
| `prelude/notify-batch` | Batch SMS notifications |
Recommendation

Require explicit user confirmation before any email/SMS send, especially batch sends, and use an API key/account with those capabilities disabled if they are not needed.

What this means

A single credential may authorize paid model calls and non-PDF actions across multiple providers through the SkillBoss gateway.

Why it was flagged

The required credential is not limited to PDF operations; the artifact says one key can access many models and providers.

Skill content
metadata: {"clawdbot":{"requires":{"env":["SKILLBOSS_API_KEY"]},"primaryEnv":"SKILLBOSS_API_KEY"}}
...
One API key, 50+ models across providers (Bedrock, OpenAI, Vertex, ElevenLabs, Replicate, Minimax, and more).
Recommendation

Use a dedicated, least-privileged key with spending limits and only the model families you actually need. Do not reuse a broad production key for this skill.

What this means

Prompts, documents, audio, images, or other content used with the skill may leave the local machine and be processed by SkillBoss and downstream providers.

Why it was flagged

The skill is explicitly built around sending requests to an external gateway that may route data to multiple downstream providers.

Skill content
**Base URL:** `https://api.heybossai.com/v1`
**Auth:** `-H "Authorization: Bearer $SKILLBOSS_API_KEY"`
...
One API key, 50+ models across providers (Bedrock, OpenAI, Vertex, ElevenLabs, Replicate, Minimax, and more).
Recommendation

Avoid sending confidential files or personal data unless you have reviewed SkillBoss and downstream provider privacy, retention, and compliance terms.

What this means

If a user obtains or creates a run.mjs helper from elsewhere, that code is not covered by this review.

Why it was flagged

Auxiliary docs reference a run.mjs helper, but the supplied manifest says no code files are present. This is not evidence of malicious behavior, but it means any such helper would be outside the reviewed artifacts.

Skill content
run.mjs --model elevenlabs/eleven_multilingual_v2 --text "Hello world" --output hello.mp3
Recommendation

Use the visible curl examples or inspect any external helper script before running it with the API key.