Versioned security assessment

Report ID: SA-C47A0D68

7/12/2026, 9:48:13 AM

video-ad-specs security assessment v1

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
video-ad-specs
Version
v1
Maintainer
101-skills
Source
github.com/101-skills/skills/tree/main/guides/video/video-ad-specs/
Coverage
1 Files scanned · 261 Lines analyzed
Policy version
Unavailable

Highest confirmed finding severity

Medium

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.

Seven command findings are confirmed because the skill directs users or agents to invoke remote CLI operations, including package installation and media generation. Markdown fence matches, ordinary reference links, and alleged reconnaissance text are false positives. A broad belt command permission is an additional medium-risk capability.

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

1 Files scanned · 261 Lines analyzed

9 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.

  1. Source

    Commit and path bound

  2. Artifact

    Content and tree hashes bound

  3. Audit

    Complete

  4. 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.

Not recorded by this audit

Env variables

May read values from the process environment.

Not recorded by this audit

External commands

May invoke commands or programs outside the Skill.

Observed in 14 evidence locations

Capability review items (8)
Medium
SKILL.md:7
Ruby/shell backtick execution
> **Install the belt CLI skill:** `npx skills add belt-sh/cli`
Line 7 instructs users to run npx against a remote skill package. This can download and execute third-party installation logic.
Medium
SKILL.md:17-26
Ruby/shell backtick execution
```bash
The fenced Bash example invokes belt login and a remote video-generation application. These are intended external CLI operations, despite the inaccurate Ruby pattern label.
Medium
SKILL.md:120-149
Ruby/shell backtick execution
```bash
This Bash block runs several remote belt applications for video, speech, merging, and captions. Following it performs external commands and processes user media.
Medium
SKILL.md:153-157
Ruby/shell backtick execution
```bash
The fenced Bash example invokes a remote belt video-generation application. This is an intended external command, not Ruby backtick syntax.
Medium
SKILL.md:161-171
Ruby/shell backtick execution
```bash
This Bash block invokes remote video-generation and text-to-speech applications through belt. Executing the example communicates with third-party services.
Medium
SKILL.md:188-195
Ruby/shell backtick execution
```bash
The Bash example runs a remote captioning application with user-supplied media paths. This is an external CLI operation with data-processing implications.
Medium
SKILL.md:252-257
Ruby/shell backtick execution
```bash
The Bash block contains four npx installation commands for remote skills. Running them can retrieve and execute third-party package logic.
Low
SKILL.md:15
Hardcoded URL
> Requires inference.sh CLI (`belt`). [Install instructions](https://raw.githubusercontent.com/infer
The installation link targets a mutable raw file on the main branch. Following changing remote installation instructions creates a supply-chain risk.

Risk findings

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

Confirmed security concerns (1)

RISK-001 Medium
SKILL.md:4
Broad External Command Permission
The allowed-tools declaration authorizes every belt subcommand. A mistaken or manipulated prompt could invoke unrelated belt operations beyond video production.
The wildcard permission is explicit on line 4 and is broader than the documented app commands. Its scope is directly observable.

Remediation

Suggested fixes recorded by this audit. Applying them is the maintainer’s responsibility.

  1. FIX-001
    Medium
    The skill recommends installing dependencies with unpinned npx commands.
    Pin each skill dependency to a reviewed version or immutable commit and require user confirmation before installation.
  2. FIX-002
    Medium
    The allowed-tools declaration permits every belt subcommand.
    Restrict allowed commands to the specific belt app operations required for video generation and processing.
  3. FIX-003
    Low
    Installation guidance points to a mutable raw file on the main branch.
    Link to a versioned release document and publish checksums or signatures for downloaded tooling.

Expert evidence

Immutable subject identity, scanner metadata, dismissed matches, and source-level evidence.

Artifact subject

Marketplace commit
c47a0d6818048b11e035c26cf0a679adb938020c
Content hash
e4097c8aa52d6dd1cb9ffbd1fe1a715fe8378a1eecd90181ea617347606677d3
Tree hash
e254f4d19d91dd4d32870d21926886edb8c5b03d4fc9c4ca73dbee92914927cd
Skill path
skills/101-skills/video-ad-specs
Audit payload hash
ad0a180bcd577ea89ba9b19a37ba85a2

Analysis metadata

Audit model: codex

Analysis state: Complete

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

Verify and export

JSON report CITATION.cff SARIF 2.1.0 Signed installation manifest Source

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

Audit attestation: not_attestable