Bug bounty AI tools
These bug bounty AI tools let you choose the model and the provider, so program data goes only where you send it. Each setup follows the tool's own docs.
| Tool | Type | Calls | Setup | In a hunt |
|---|---|---|---|---|
| Claude Code | Terminal agent | Messages | Our steps, docs | Sorts a notes folder, writes parsers for your own tool output, drafts reports from files. |
| Codex CLI | Terminal agent | Responses | Our steps, docs | The same jobs as Claude Code, configured in one TOML file. |
| OpenCode | Terminal agent | Chat Completions | Our steps, docs | Open-source agent that takes any OpenAI-compatible provider. |
| Open WebUI | Chat app you host | Chat Completions | Our steps, docs | A private chat window for scope questions and report drafts. |
| LLM CLI | Command line | Chat Completions | Our steps, docs | Pipes a notes file into a model from the shell. |
| Cline | VS Code extension | Chat Completions | docs | API Provider "OpenAI Compatible", then Base URL, API Key and Model ID. |
| Caido | Web proxy | Custom providers | docs | Settings > AI > Providers on Team and Enterprise plans, from v0.59.0. |
| Burp Suite | Web proxy | None for the built-in AI | docs | Its docs show no setting for your own model in the built-in AI. |
Read November 18, 2026 in each tool's own docs.
Caido
"the Assistant does not support API keys for other AI providers or configuration for use with local LLMs."
Caido docs, Assistant model, read November 18, 2026
Release 0.59.0 added custom providers under Settings, AI, Providers on Team and Enterprise plans (release notes).
Burp Suite
PortSwigger's Burp AI docs show no setting for your own model. Use a separate tool from this list for report drafts.
Every tool here needs
| Value | Example |
|---|---|
| Base URL | https://api.silkcompute.com, or with /v1 where the tool asks |
| API key | Issued by the provider, kept in an environment variable |
| Model id | kimi-k3.1-pro-uncensored |