Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent spec-brainstorming skill, but it can inspect the workspace and create a git commit without prominent, separate user consent.
Install only if you are comfortable with a brainstorming skill reading broad repository context and saving specs under .beagle. Before use, treat any git commit as requiring explicit approval: ask the agent to show the target path and diff, and to skip committing unless you specifically request it.
The skill claims it only produces a spec and does not take implementation action, but later instructs the agent to write files to disk and create a git commit. That mismatch can mislead users and higher-level orchestrators about the skill's side effects, enabling unexpected state changes in the workspace and repository.
Automatically committing to git is a high-impact side effect that is not necessary for brainstorming/spec generation and can permanently alter project history. If invoked unexpectedly or on sensitive repositories, it could create unwanted commits, leak generated content into version control, or interfere with existing workflows and approvals.
The trigger phrases are broad enough that the skill may activate for generic brainstorming or project-help requests outside its intended scope. Because this skill can inspect the repository, write files, and even commit to git, overbroad activation increases the chance of unintended side effects from an imprecise user request.
The stated purpose is a dialogue-centered brainstorming/spec skill that does not plan or implement. Yet the workflow includes reading project files, docs, git history, running a repo-wide prior-art sweep, and potentially invoking web-research or artifact-analysis, which materially expands the behavior from conversational clarification into repository investigation and external research orchestration.
The skill description presents the tool as producing a spec through dialogue, but does not upfront warn that it may write files and persist artifacts. Missing disclosure undermines informed consent and can cause users to invoke the skill without realizing it will modify the workspace.
Committing to git without a clear user-facing warning is especially dangerous because it creates durable repository state and may trigger downstream automation, review noise, or accidental disclosure through later pushes. The absence of prominent disclosure makes the side effect surprising and harder for users to control.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
**Only fix issues that would cause real problems downstream.**
A downstream planning system acting on this spec should be able to:
- Understand what to build without asking the user again
- Decompose requirements into tasks
- Know what's in scope and what's not
- Understand the constraints they're working within
No suspicious patterns detected.