# AI use cases for Skills (zero-lab-ai/skills) Repo: [zero-lab-ai/skills](https://github.com/zero-lab-ai/skills) | Use cases: 4 | Page: https://dedupedai.com/use-cases/zero-lab-ai-skills/ A collection of 56 portable security skills in 10 installable plugins for authorized defensive assessment, red teaming, penetration testing, detection engineering, and engagement operations. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Audit agent tools for poisoning and excess privilege: https://dedupedai.com/use-cases/zero-lab-ai-skills/agent-tool-surface-audit/ - Test RAG tenant boundaries with controlled fixtures: https://dedupedai.com/use-cases/zero-lab-ai-skills/rag-tenant-boundary-check/ - Plan an authorized engagement and assemble findings: https://dedupedai.com/use-cases/zero-lab-ai-skills/engagement-plan-and-findings/ - Review AD certificate template exposure from exports: https://dedupedai.com/use-cases/zero-lab-ai-skills/adcs-template-exposure-review/ ## How to audit agent tools for poisoning and excess privilege with Skills Repo: [zero-lab-ai/skills](https://github.com/zero-lab-ai/skills) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Review an agent's tool definitions for poisoning, excess privilege, and deployment blast radius before rollout. **Who it is for:** An AI platform engineer about to give an agent new tools who needs a review before the agent goes live. **The problem:** Agent tools can carry poisoned instructions, unsafe parameters, or more privilege than the task needs. **The outcome:** A ranked list of tool-surface findings, each tied to the tool definition it came from. ### Steps 1. Install the plugin marketplace in Claude Code with `/plugin marketplace add zero-lab-ai/skills`. 2. Run `/plugin menu` and select the agent-security bundle. 3. Give the agent-surface-audit skill the tool definitions and deployment context you are authorized to review. 4. Check each finding against the tool's real permissions before changing any configuration. ### Prompt to paste into your agent ```text Using the agent-surface-audit skill from the zero-lab-ai/skills repo, review the tool definitions for that I am authorized to assess. Return findings for poisoning, excess privilege, unsafe parameters, and deployment blast radius, each with the tool definition it cites. Verify that every finding quotes a definition I provided. ``` ### Caveat The Skills support authorized work and do not grant permission or replace your organization's rules of engagement. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/zero-lab-ai-skills/agent-tool-surface-audit/ ## How to test RAG tenant boundaries with controlled fixtures with Skills Repo: [zero-lab-ai/skills](https://github.com/zero-lab-ai/skills) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Measure whether one tenant's documents can surface for another tenant using benign fixtures you control. **Who it is for:** A developer or security reviewer on a multi-tenant retrieval app who needs to confirm tenant isolation before a release. **The problem:** Retrieval filters can fail quietly, so one tenant's content may appear in another tenant's answers. **The outcome:** A record of exposure states for each tenant boundary tested, with the fixtures used. ### Steps 1. Install the plugin marketplace in Claude Code with `/plugin marketplace add zero-lab-ai/skills`, or in Codex with `codex plugin marketplace add zero-lab-ai/skills`. 2. Select the rag-security bundle from the plugin menu. 3. Create benign canary documents in test tenants that you own, and record which tenant each one belongs to. 4. Run the rag-data-exposure-audit skill against the retrieval endpoint and record each exposure state it reports. ### Prompt to paste into your agent ```text Using the rag-data-exposure-audit skill from the zero-lab-ai/skills repo, test the retrieval endpoint with the canary fixtures I created in tenants and . Return one exposure state per boundary, with the query and the returned canary. Verify that every canary returned belongs to the tenant named in the result. ``` ### Caveat Run this only against tenants you own or have written permission to test, because the Skills do not expand scope. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/zero-lab-ai-skills/rag-tenant-boundary-check/ ## How to plan an authorized engagement and assemble findings with Skills Repo: [zero-lab-ai/skills](https://github.com/zero-lab-ai/skills) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Turn your rules of engagement into a plan, then build evidence-backed findings and a report draft. **Who it is for:** A penetration tester or engagement lead starting a new authorized assessment who needs a structured plan and report trail. **The problem:** Engagement notes, evidence, and findings often drift apart, which weakens the final report. **The outcome:** A plan tied to your rules of engagement, plus findings that each point to their supporting evidence. ### Steps 1. Install the plugin marketplace in Claude Code with `/plugin marketplace add zero-lab-ai/skills`. 2. Run `/plugin menu` and select the engagement-operations bundle. 3. Paste your signed rules of engagement and scope list into the session, and mark anything out of scope. 4. Record each finding with the evidence item that supports it before writing the report section. ### Prompt to paste into your agent ```text Using the engagement-operations skills from the zero-lab-ai/skills repo, draft an engagement plan for based only on the rules of engagement I provide. Return the plan, a findings list with evidence references, and a report outline. Verify that no target or action appears that is outside the scope I gave. ``` ### Caveat The plan is only as good as the rules of engagement you supply, since the Skills do not expand scope on their own. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/zero-lab-ai-skills/engagement-plan-and-findings/ ## How to review AD certificate template exposure from exports with Skills Repo: [zero-lab-ai/skills](https://github.com/zero-lab-ai/skills) | Difficulty: Advanced | Time: 1 hour | Basis: From the README Audit certificate templates, CA policy, and enrollment controls from supplied directory exports. **Who it is for:** An Active Directory administrator preparing a PKI review who has authorized access to certificate template and CA policy exports. **The problem:** Weak certificate template settings or permissive enrollment rights can create paths to privileged identities. **The outcome:** A list of template and CA policy controls rated by exposure, with the export line behind each rating. ### Steps 1. Install the plugin marketplace in Codex with `codex plugin marketplace add zero-lab-ai/skills`. 2. Run `codex plugin list` to confirm the identity-security bundle is available. 3. Export certificate template settings, CA policy, and PKI ACLs that you are authorized to review. 4. Provide the exports to the adcs-exposure-audit skill and review each control it flags against the export. ### Prompt to paste into your agent ```text Using the adcs-exposure-audit skill from the zero-lab-ai/skills repo, audit the certificate template and CA policy exports for that I provide. Return each control with an exposure rating and the export line it cites. Verify that every rating cites a line from the exports I supplied. ``` ### Caveat The skill works from evidence you supply, so it cannot confirm live exposure without your own validation under the engagement's rules. Risk, credentials: Keep private keys, passwords, and tokens out of the prompt, and share only the template and policy settings you are authorized to review. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/zero-lab-ai-skills/adcs-template-exposure-review/