Use this plugin when you need a structured workflow rather than a one-off answer. Its capabilities are organized into focused skills so you can start broad, then go deeper where the work requires it.
Injection Checker
Agent Tool Boundaries
Indirect Injection Review
Mitigation Regression Tests
Use cases
→Ask Prompt Injection Security to apply injection checker to a real project, document, or decision and explain the assumptions it used.
→Ask Prompt Injection Security to apply agent tool boundaries to a real project, document, or decision and explain the assumptions it used.
→Ask Prompt Injection Security to apply indirect injection review to a real project, document, or decision and explain the assumptions it used.
→Ask Prompt Injection Security to apply mitigation regression tests to a real project, document, or decision and explain the assumptions it used.
Skills included
injection checker
Orchestrate evidence-based prompt-injection risk reviews across prompts, retrieval, multimodal input, memory, and agent tools.
What it helps with
Use Injection Checker to turn a focused Prompt Injection Security task into a clearer, repeatable workflow.
Example use
Ask Prompt Injection Security to apply injection checker to a real project, document, or decision and explain the assumptions it used.
Prompt pattern
Use Injection Checker for [specific goal]. Context: [relevant files, audience, inputs, or constraints]. Return: [desired format] and flag uncertainties.
agent tool boundaries
Review AI agent tools, permissions, arguments, outputs, and action approvals for prompt-injection-driven misuse paths.
What it helps with
Use Agent Tool Boundaries to turn a focused Prompt Injection Security task into a clearer, repeatable workflow.
Example use
Ask Prompt Injection Security to apply agent tool boundaries to a real project, document, or decision and explain the assumptions it used.
Prompt pattern
Use Agent Tool Boundaries for [specific goal]. Context: [relevant files, audience, inputs, or constraints]. Return: [desired format] and flag uncertainties.
indirect injection review
Examine documents, web pages, retrieval chunks, messages, and multimodal inputs as possible indirect prompt-injection sources.
What it helps with
Use Indirect Injection Review to turn a focused Prompt Injection Security task into a clearer, repeatable workflow.
Example use
Ask Prompt Injection Security to apply indirect injection review to a real project, document, or decision and explain the assumptions it used.
Prompt pattern
Use Indirect Injection Review for [specific goal]. Context: [relevant files, audience, inputs, or constraints]. Return: [desired format] and flag uncertainties.
mitigation regression tests
Turn prompt-injection findings into defense-in-depth fixes and safe repeatable regression tests.
What it helps with
Use Mitigation Regression Tests to turn a focused Prompt Injection Security task into a clearer, repeatable workflow.
Example use
Ask Prompt Injection Security to apply mitigation regression tests to a real project, document, or decision and explain the assumptions it used.
Prompt pattern
Use Mitigation Regression Tests for [specific goal]. Context: [relevant files, audience, inputs, or constraints]. Return: [desired format] and flag uncertainties.
How to use Prompt Injection Security
1Start with a specific outcome and provide the relevant context.
2Ask the plugin to show its assumptions and structure the result.
3Review the first result, add missing constraints, and request a focused revision.
4Save the final output in the format your workflow needs.
Example prompts
Help me use Prompt Injection Security to achieve [goal]. Context: [details]. Return [format] and list assumptions.
Apply the Prompt Injection Security workflow to this input: [paste input]. First explain your approach, then produce the result.
Review this result from Prompt Injection Security against these criteria: [criteria]. Recommend the smallest improvements.
How to get maximum benefit
Name the audience, goal, inputs, constraints, and output format.
Give the plugin one clear job per request before expanding scope.
Ask it to identify uncertainty rather than filling gaps with guesses.
Review important outputs before publishing or acting on them.
Limitations and safe use
Results depend on the quality and completeness of the context provided.
Review important factual, technical, financial, legal, medical, or public-facing outputs before use.
The plugin should support decisions and workflows, not replace qualified human judgment.
Frequently asked questions
What is Prompt Injection Security?
Prompt Injection Security is a security plugin created by The Doers Firm. It includes 4 focused skills for developers and teams shipping applications.
How do I get the best results from Prompt Injection Security?
Give it a specific goal, relevant context, constraints, and a desired output format. Ask it to surface assumptions and revise the result against clear criteria.
