Skip to content
skillsafebeta
← Report library
Purpose / Other

Implement Skill Security Audit

What the author says it does (original text)

Implement a piece of work based on a spec or set of tickets.

Independent security check

Security risks found

Files checked
2
Risks found
2
Could it run dangerous commands?Looks for programs run straight after downloading, remote control of your computer, and hidden commands.Risks found: 1
Medium risk

The full test suite may execute repository-controlled code and hooks

Source references: 3
What we found

The Skill requires a final full-suite run. Test commands commonly load repository configuration, plugins, fixtures, and lifecycle scripts, while this instruction sets no trust boundary, exact command, network restriction, or side-effect limit.

Why this matters

In an untrusted or compromised repository, the test path could read available files or credentials, use the network, start services, or modify the workspace or external resources. The supplied evidence does not show that this will occur, but the instruction activates that execution path.

Legitimate use of this code

The source requests type checking and tests as normal verification for implementing work from a specification. No test command, hook, network operation, or suspicious script is provided, so these lines do not support the more specific claim that repository hooks will run. For an untrusted repository, the user can still require disclosure of the exact test command and its script definition before execution.

This assessment concerns the code and conditions shown, not proof that harm has occurred.
SKILL.md:9In the instructionsOpen original file
Use /tdd where possible, at pre-agreed seams.Run typechecking regularly, single test files regularly, and the full test suite once at the end.
Show 2 other places
SKILL.md:2In the instructionsOpen original file
---name: implementdescription: "Implement a piece of work based on a spec or set of tickets."disable-model-invocation: true
SKILL.md:7In the instructionsOpen original file
Implement the work described by the user in the spec or tickets.Use /tdd where possible, at pre-agreed seams.Run typechecking regularly, single test files regularly, and the full test suite once at the end.
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

The Skill automatically creates a commit on the current branch

Source references: 1
What we found

It unconditionally says to “Commit your work,” without requiring confirmation of the commit scope, destination branch, or whether the worktree already contains user changes.

Why this matters

If the agent stages files outside this task, existing user changes could be included. Even with correct staging, it creates persistent branch history that can affect review, rebasing, or release workflows.

This is an active, unconditional instruction to create a commit. If the worktree contains pre-existing user changes or the current branch is not the intended target, the agent could record out-of-scope content in Git history. The user can ask the author to require a staged-scope review, include only task-specific files, and avoid committing without explicit authorization.

SKILL.md:13In the instructionsOpen original file
Once done, use /code-review to review the work.Commit your work to the current branch.
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

0 instruction sections

The Skill’s main behavior is to implement code from the user’s specification or tickets, with regular type checks and targeted tests followed by one full test-suite run.

View source
SKILL.md:7In the instructionsOpen original file
Implement the work described by the user in the spec or tickets.Use /tdd where possible, at pre-agreed seams.Run typechecking regularly, single test files regularly, and the full test suite once at the end.

After implementation, it invokes another capability named /code-review and requires committing the work to the current Git branch. The supplied source does not include the implementation of /tdd or /code-review, so their additional behavior cannot be determined from this evidence.

View source
SKILL.md:13In the instructionsOpen original file
Once done, use /code-review to review the work.Commit your work to the current branch.

The configuration disables implicit invocation, so the Skill is intended to run only when explicitly selected.

View source
SKILL.md:2In the instructionsOpen original file
---name: implementdescription: "Implement a piece of work based on a spec or set of tickets."disable-model-invocation: true---
agents/openai.yaml:4In the instructionsOpen original file
  short_description: "Build work from a spec or tickets"policy:  allow_implicit_invocation: false
Start here · InstructionsSKILL.md
implement

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
22
File checksum (to compare versions)
bc98e00cac258c2f67013a8852ed91300eb41f1daf1f8522248abd5d65a6c435