Skip to content
skillsafebeta
← Report library
Purpose / Writing

Gws Modelarmor Skill Security Audit

What the author says it does (original text)

Google Model Armor: Filter user-generated content for safety.

Independent security check

Security risks found

Files checked
1
Risks found
1
Could it run dangerous commands?Looks for programs run straight after downloading, remote control of your computer, and hidden commands.No risks found
Could it expose your files or keys?Looks for uploads of files containing passwords or keys, and keys written directly in the code.No risks found
Could it delete files or keep running?Looks for broad file deletion, disk overwrites, and programs set to start automatically.Risks found: 1
Medium risk

A missing dependency triggers a generator that creates additional Skill files

Source references: 1
What we found

The prerequisite explicitly directs the agent to run `gws generate-skills` when the neighboring shared Skill is absent. This goes beyond inspecting the current Skill: it executes a local binary and generates instructions that were not supplied for this audit.

Why this matters

Installing this Skill could lead the agent to modify the Skills directory merely because a dependency is missing, then follow newly generated authentication and security instructions. Their exact contents and file scope cannot be verified from the supplied source.

This is an active prerequisite, not an example or warning: if the shared Skill is missing, it directs the agent to run the local `gws generate-skills` command. That may create files and instructions not included in this audit, while the source does not state the generation scope, overwrite behavior, or whether user confirmation is required. “Automatically” is slightly overstated, but the conditional filesystem-change risk is supported. Users can ask the author to list affected paths and require explicit approval before execution.

SKILL.md:16In the instructionsOpen original file
> **PREREQUISITE:** Read `../gws-shared/SKILL.md` for auth, global flags, and security rules. If missing, run `gws generate-skills` to create it.
Could it bypass safety checks?Looks for skipped website security checks, excessive file access, or actions that skip your approval.No risks found
Could it mislead the AI or hide text?Checks the skill instructions for requests to ignore you, influence the report, or hide text in invisible characters.No risks found
Could it change links or payment recipients without asking?Looks for forced referral or payment changes combined with instructions to hide the change.No risks found

Inside this skill

2 instruction sections

This Skill is a command index for the Google Model Armor CLI. It tells the agent to inspect help and the method schema before calling an API, then construct parameters from that output. The shown commands only perform command discovery; they do not specify content, a project, or an API operation.

View source
SKILL.md:32In the instructionsOpen original file
Before calling any API method, inspect it:```bash# Browse resources and methodsgws modelarmor --help# Inspect a method's required params, types, and defaultsgws schema modelarmor.<resource>.<method>```Use `gws schema` output to build your `--params` and `--json` flags.

Authentication, security rules, and the three helper workflows are defined by neighboring Skills that are not included in the supplied source. This file therefore does not establish how credentials are used, where content is sent, or what cloud-side changes template creation performs.

View source
SKILL.md:16In the instructionsOpen original file
> **PREREQUISITE:** Read `../gws-shared/SKILL.md` for auth, global flags, and security rules. If missing, run `gws generate-skills` to create it.
SKILL.md:26In the instructionsOpen original file
|---------|-------------|| [`+sanitize-prompt`](../gws-modelarmor-sanitize-prompt/SKILL.md) | Sanitize a user prompt through a Model Armor template || [`+sanitize-response`](../gws-modelarmor-sanitize-response/SKILL.md) | Sanitize a model response through a Model Armor template || [`+create-template`](../gws-modelarmor-create-template/SKILL.md) | Create a new Model Armor template |
Start here · InstructionsSKILL.md
gws-modelarmor
Lines connect the instruction file to its sections, not an observed execution order. Select a section to read the source.
Files and check records1 files

Coverage and gaps

Content covered in each file

These are the source ranges included in this check, not a guarantee that every issue has been resolved.

  • SKILL.mdFull text included

This report is for the version above. We read the available code and instructions without running the skill or checking extra packages it installs. This is not a promise of safety: a different version or setup may behave differently.

  • SKILL.mdInstructions

Operations mentioned in code and instructions

Run commands
SKILL.md:18In the instructionsOpen original file
```bashgws modelarmor <resource> <method> [flags]
SKILL.md:34In the instructionsOpen original file
```bash# Browse resources and methods
Lines read
44
File checksum (to compare versions)
817f512cf10195a2cf2c4cd4b5bfa880026b08ad68626d8d39bc35f6a709d6a6