Set up Cline with PentestLLM
Cline connects to PentestLLM through its OpenAI Compatible provider: open Cline’s settings, choose that provider, and enter the base URL https://api.pentestllm.ai/v1, your key and the model id auto/security. Nothing else in the editor changes, and Cline’s approval prompts stay in force.
https://api.pentestllm.ai/v1
Steps
Open Cline’s settings
In the Cline panel, click the gear icon. This opens the settings where the API provider is chosen.
Choose the OpenAI Compatible provider
In the API Provider list, select OpenAI Compatible. This is the provider Cline documents for endpoints other than the official OpenAI API.
Enter the base URL and key
Paste the PentestLLM base URL into Base URL, including the
/v1, and your key from the console into API Key.Base URL https://api.pentestllm.ai/v1Enter the model id
Type the model id into the Model field. The router id works here as written; a single build id from the models page works the same way.
Model auto/securitySet the model configuration
Open Model Configuration and set Context Window size and Max Output Tokens to the figures on the build’s model page, so Cline knows when to trim the conversation. If you pinned
gemma-4-26b-a4b-it-cybersecurity, its context window in the catalog is 262,144 tokens. Input Price and Output Price are optional; filling them from the pricing page makes Cline’s cost display accurate.
What each field does
- API Provider
- OpenAI Compatible. Tells Cline to send the OpenAI chat format to a URL you name.
- Base URL
https://api.pentestllm.ai/v1. Cline’s docs note it must not be the official OpenAI URL.- API Key
- Your PentestLLM key, sent as a bearer token.
- Model
auto/security, or one build id exactly as the models page shows it.- Context Window size
- How much conversation Cline keeps before trimming. Use the build’s real window.
- Max Output Tokens
- The cap Cline requests per answer.
- Input Price, Output Price
- Only used for Cline’s own cost readout. Billing happens on the API side.
If something fails
- Invalid API key
- Re-copy the key from the console and check there is no space around it.
- Model not found
- The id is not one the API serves. Copy it from the models page; display names do not work.
- Connection errors
- Check the base URL ends in
/v1and that your network can reach it. - Answers stop mid-task
- The balance may be empty. A top-up in the console resumes service.
These are the same checks Cline’s own troubleshooting list gives for any OpenAI-compatible provider. The curl command below tests the key and URL without the editor.
curl https://api.pentestllm.ai/v1/models -H "Authorization: Bearer sk-ptl-live-…"Why the router suits Cline
Cline works in a loop: it reads files, proposes an edit or a command, waits for your approval and continues. Requests in that loop vary from a few hundred tokens to tens of thousands. With auto/security, turns that carry tools and code go to a stronger build and short turns go to a fast one, each billed at its own price. If you want one model for the whole task, enter a single build id instead.
Security review inside the editor
On a security build, Cline becomes a code reviewer that does not decline the subject: ask it to read a module for injection or access control flaws, explain a dependency’s advisory, or draft the finding. The builds answer security questions other models decline; refusals are reduced, not gone, and its findings are leads to confirm.
Leave Cline’s approval prompts on for commands. PentestLLM is for code and systems you own or have written permission to test, per the acceptable use policy. For a terminal agent, see the OpenCode guide; for prices, the pricing page.
9 security builds, priced per token
Read from the live catalog on 2026-11-11. Prices are US dollars per million tokens, paid from a prepaid balance that starts at $10. The id auto/security has no price of its own: each request is billed at the build that answered.
| Build | Input | Output | Context |
|---|---|---|---|
| Qwen3.8 27B Cybersecurity | $0.40 | $2.40 | 262,144 |
| Gemma 4 26B A4B Cybersecurity | $0.4224 | $1.3376 | 262,144 |
| GLM-5.3 Flash Cybersecurity | $0.5093 | $1.6975 | 1,048,576 |
| Red Team (Flash) | $0.90 | $3.00 | not published |
| Red Team (Pro) | $1.50 | $6.00 | not published |
| OrcaCyber MiniMax M3.1 Flash | $2.00 | $4.00 | 261,632 |
| OrcaCyber Qwen3.8 Flash Next | $2.00 | $4.00 | 261,632 |
| Cyber Frontier | $5.25 | $15.75 | 1,048,576 |
| OrcaCyber Zero 1.5 | $7.50 | $18.75 | 999,488 |
Questions
OpenAI Compatible. It takes a base URL, an API key and a model id, which are the three values PentestLLM needs.
Enter https://api.pentestllm.ai/v1 in the Base URL field, including the /v1 at the end.
Yes. Type auto/security in the Model field and the API chooses a security build for each request and names it in the response.
Cline uses that figure to decide when to trim the conversation. If it is larger than the build’s real window, long tasks fail; if it is smaller, Cline forgets earlier steps sooner than it needs to.
Per token from a prepaid balance. On 2026-11-11 input prices for security builds start at $0.40 per million tokens, and the minimum top-up is $10.
Read next
- All agent guidesThe same three values in every supported client.
- CursorThe OpenAI key and base URL override in Cursor Settings, and its limits.
- OpenCodeThe opencode.json provider block for the security builds.
- Bug bountyModels for scope reading, recon notes, code review and report drafting on programs you are enrolled in.
- Vulnerability researchModels for code review, CVE and patch analysis, and advisory writing.
Facts about other products and tools on this page were read from their own pages on 11 November 2026: Cline docs: OpenAI Compatible. They change; check the source before relying on a figure.
One key, from $10
PentestLLM is for systems you own or have written permission to test. The acceptable use policy says what is not allowed.
- Input, 8,000 tokens at $0.40 per million
- $0.0032
- Output, 1,000 tokens at $2.40 per million
- $0.0024
- One turn
- $0.0056
- 1,000 turns
- $5.60