Guide · Agent skill
How to use the Managed Pentesting with Strix agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
This guide explains the managed-pentesting-with-strix agent skill, which lets an AI agent run Strix's autonomous penetration testing on the Strix cloud platform at app.strix.ai instead of on your own machine. It is written for readers who have not used agent skills before. By the end you will know how to install the skill, how to ask an agent for a managed scan in plain language, and which safety and plan limits apply.
Managed Pentesting with Strixusestrix/strix · Apache-2.0 Download the skill Quick start
- 1Download and unzip the skill. Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder.
- 2Install the strix cloud CLI. Install the CLI using the curl command from strix.ai/install, then confirm it runs.
- 3Sign in with least-privilege scopes. Run strix cloud login with the scopes the task needs and approve the sign-in in your browser.
- 4Check your credits. Run strix cloud credits to confirm you have enough balance for a non-Enterprise scan.
- 5Ask the agent for a managed scan. Describe the target and the goal in plain language, and let the agent register the asset, start the scan and poll its progress.
- 6Review and export findings. Ask the agent to triage vulnerabilities and export SARIF or a compliance report if your plan allows it.
What does this skill make the agent do?
A skill is a folder containing a SKILL.md file that tells an AI agent how to perform a specific task. This skill teaches the agent to use the strix cloud CLI or the Strix REST API to run security tests on Strix's platform. The scan runs on Strix's infrastructure, so you do not need local Docker or your own LLM key.
The agent can sign in, register web apps, APIs, repositories or local source, start and poll scans, review and update vulnerability findings, export SARIF, download compliance reports, trigger pull request security reviews, and set up schedules or webhooks. Results are tracked in a team dashboard.
- Best for recurring or scheduled testing that a team tracks in a shared dashboard.
- Best for hosted agent or CI environments where Docker is unavailable.
- Not for systems you do not own or have written permission to test. The skill does not verify authorisation.
- Not for air-gapped or offline work, and not for cases where source code must stay off third-party servers, because local source is uploaded to the platform.
How do you install the skill?
Download the skill ZIP from this page and unzip it. For a personal install, place the folder in ~/.claude/skills/. For a single project, place it in that project's .claude/skills/ folder. In the Claude apps you can instead upload the skill under Settings, then Capabilities, then Skills.
The skill also needs the strix cloud CLI, which the source says is installed with a curl command from strix.ai/install. Check the install page for the exact command before running it.
- Personal install: ~/.claude/skills/managed-pentesting-with-strix/
- Project install: .claude/skills/managed-pentesting-with-strix/ inside your project
- Claude apps: Settings, then Capabilities, then Skills, then upload
How do you sign in and set scopes?
Before the agent can do anything, you sign in with the device flow. The CLI creates your account and workspace on first use and stores a personal API token in ~/.strix/platform-auth.json. You approve the sign-in in your browser.
Use least-privilege scopes. For a non-interactive agent shell, pass the scopes explicitly so no terminal prompts appear. Never hardcode, log or commit the token. Store it in an environment variable or your CI secret store.
- strix cloud whoami shows the fast local status.
- strix cloud session --json verifies the remote session.
- strix cloud logout revokes the remote session before removing the local token.
strix cloud login --scopes scans:read scans:write uploads:write billing:read vulnerabilities:read assets:read assets:writeHow do you trigger the skill with natural requests?
You do not need to type commands yourself. Describe the outcome you want and name the target and the kind of test. The agent should match the request to this skill because it covers managed, continuous, scheduled, team-tracked or sandboxed-agent security testing.
Good requests name a target and a goal, for example asking the agent to run a managed scan of a staging web app and summarise the findings, or to export the latest results as SARIF for your code scanning tool.
- Run a managed pentest of our staging web app and list the high severity findings.
- Register this repository on Strix and start a code review scan.
- Set up a weekly scheduled scan for the public API and send failures to our webhook.
- Download the compliance report for the last completed scan.
What do scan results and errors look like?
Output is complete JSON when stdout is not a terminal, or when you pass --json. Paginated lists print the next --page or --offset value. Binary downloads need explicit redirection or the --output FILE --json form.
Exit codes tell the agent what went wrong. Code 0 means success, 1 is a request or runtime error, 2 is invalid usage, 4 is an authentication or plan limit, and 5 means payment is required.
- 401: bad or expired token (exit 4)
- 402: out of credits (exit 5)
- 403: scope or plan-tier limit (exit 4)
- 422: validation error (exit 1)
What does scanning cost and what plan do you need?
Non-Enterprise scans consume organisation credits. Enterprise engagements are included in the plan and do not debit the credit wallet. Check your balance before a scan with the credits command, which needs the billing:read scope.
Downloadable PDF and DOCX reports need the Enterprise plan. Schedules need the Pro plan. Billing writes, such as buying credits, need an admin token.
- strix cloud credits shows the balance.
strix cloud creditsWhat are the safety notes?
Scans send traffic to your targets, so only test systems you own or have written permission to test. The skill does not verify this for you. Local source code is uploaded to the platform, and scans need network access.
Create narrow automation tokens with an expiry, and restrict them to specific targets, tags or business units where possible. The token secret is shown only once, so store it directly in a secret manager.
- Use time-limited tokens and rotate them regularly.
- Assign only the scopes the integration needs.
- Do not print or commit token secrets.
What people use it for
- Running a recurring managed scan of a staging web app and tracking results in a shared dashboard.
- Triggering a pull request security review from a CI pipeline where Docker is unavailable.
- Exporting SARIF results to feed an existing code scanning workflow.
- Setting up a scheduled scan with a webhook that alerts your team when findings change.
Frequently asked questions
Is the Strix managed pentesting skill free?
Non-Enterprise scans consume credits on the Strix platform, so they are not free. Enterprise engagements are included in the plan and do not use the credit wallet. The open-source local skill, penetration-testing-with-strix, is described in the source as free and runs on your own machine.
Do I need Docker or an LLM key to use this skill?
No. The skill is designed to run scans on Strix's cloud, so local Docker and your own LLM key are not needed. You do still need the strix cloud CLI or REST API access and a Strix platform account.
What is the difference between this skill and the local Strix skill?
This skill runs scans on app.strix.ai and tracks results in a team dashboard. The local skill, penetration-testing-with-strix, runs on your machine and suits fully local, air-gapped or bring-your-own-LLM work. Both use the same engine and SARIF output, so you can mix them.
Why does my scan command return exit code 4 or 5?
Exit code 4 means an authentication problem or a plan limit, such as an expired token or a feature your plan does not include. Exit code 5 means payment is required, usually because your credits are used up. Check your token with strix cloud session and your balance with strix cloud credits.
Can I use the skill without a browser sign-in?
Yes, for automation. You can pass --scopes to strix cloud login so no terminal prompts appear, or pass a token through the --token option or the STRIX_API_TOKEN environment variable. Store tokens in a secret store and never commit them.
Does the skill check that I am allowed to test a target?
No. The source states that the skill does not verify authorisation, and scans send traffic to your targets. Only run it against systems you own or have written permission to test.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if your team wants recurring, scheduled or shared security testing and you are happy for scans and local source to run on Strix's platform. Pick the local Strix skill instead if you need offline work, want to keep source code and target data on your own infrastructure, or prefer no credit costs.
Download the skill