Skip to content
skillsafebeta
← Report library
Purpose / Other

Grill Me Skill Security Audit

What the author says it does (original text)

A relentless interview to sharpen a plan or design.

Independent security check

Security risks found

Files checked
2
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.No risks found
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.Risks found: 1
Medium risk

Operational instructions are delegated to an external Skill whose source is absent

Source references: 3
What we found

The sole runtime instruction calls “grilling,” but that Skill’s implementation is not provided. The displayed plan-interview purpose neither constrains nor demonstrates what the invoked Skill actually does.

Why this matters

After the user explicitly starts this wrapper, the agent may follow all instructions supplied by the environment’s current “grilling” Skill. Those instructions could request additional information or perform actions that this audit cannot verify. The supplied material does not establish which files, data, accounts, or tools it may access.

What this evidence establishes

The package’s only operational instruction delegates to another Skill named “grilling,” whose source is absent, so its file, network, or account behavior cannot be verified. The metadata indicates it should require explicit invocation, reducing accidental activation, but this does not constrain the delegated Skill. A user can ask for the dependency’s complete, pinned source and restrict its file, network, and account permissions before use.

This assessment concerns the code and conditions shown, not proof that harm has occurred.
SKILL.md:2In the instructionsOpen original file
---name: grill-medescription: A relentless interview to sharpen a plan or design.disable-model-invocation: true---Call the Skill tool with "grilling".
Show 2 other places
agents/openai.yaml:4In the instructionsOpen original file
  short_description: "Sharpen a plan through interview"policy:  allow_implicit_invocation: false
SKILL.md:7In the instructionsOpen original file
Call the Skill tool with "grilling".
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

0 instruction sections

This Skill does not implement the interview itself; when run, it directs the agent to call another Skill named “grilling.” Its actual behavior therefore depends on whichever external Skill that name resolves to in the current environment, whose contents are not included here.

View source
SKILL.md:2In the instructionsOpen original file
---name: grill-medescription: A relentless interview to sharpen a plan or design.disable-model-invocation: true---Call the Skill tool with "grilling".

The configuration disallows implicit model invocation, so the Skill ordinarily requires explicit selection by the user or system before it starts.

View source
SKILL.md:4In the instructionsOpen original file
description: A relentless interview to sharpen a plan or design.disable-model-invocation: true---
agents/openai.yaml:4In the instructionsOpen original file
  short_description: "Sharpen a plan through interview"policy:  allow_implicit_invocation: false
Start here · InstructionsSKILL.md
grill-me

The instructions have no section headings. Supporting files and code excerpts are available below.

Lines connect the instruction file to its sections, not an observed execution order. Select a section to read the source.
Files and check records2 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
  • agents/openai.yamlFull 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
  • agents/openai.yamlSupporting file
Lines read
14
File checksum (to compare versions)
a8d7ef7e0db48a449fd2dce4031f8f3e434c2b7c3c9de0efc8ffe2e3f644c967