bounty-operator

<div align="center">

# Bounty Operator

**Find the hole in your report before the triager does.**

Bounty Operator argues against your bug bounty report before you submit it.
Give it the draft and the code it cites. It checks every claim against the
code, the way a triager will, and you get one verdict: `submit`,
`rewrite-then-submit`, `prove-first`, `hold-duplicate` or `drop`. It runs on
your own model: in the browser, in your coding agent over MCP, or from the
command line.

### [Start a free review at bountyoperator.com](https://bountyoperator.com/)

[![Release](https://img.shields.io/github/v/release/bountyoperator/bounty-operator?sort=semver)](https://github.com/bountyoperator/bounty-operator/releases)
[![License](https://img.shields.io/github/license/bountyoperator/bounty-operator)](LICENSE)
![Python](https://img.shields.io/badge/python-3.10%2B-blue)

<a href="https://bountyoperator.com/"><img src="docs/assets/review.png" alt="A Bounty Operator review of a draft report: the verdict, then each claim marked confirmed, overstated or contradicted against file and line" width="820"></a>

</div>

- **Every claim tied to file and line.** Findings cite `input-1/src/Vault.sol:142-158`.
  A reference to a file or line you did not supply is flagged. Each review
  downloads as a packet with a SHA-256 manifest of exactly what was reviewed.
- **The triager's objection first.** Each finding carries its strongest
  counterargument, whether the source resolves it, and the one missing artifact
  that would settle it.
- **Your model and your key. Nothing is stored.** OpenRouter, Anthropic, OpenAI,
  Gemini, xAI, DeepSeek, Mistral or Groq with your own key. A hosted review
  passes your files and your key through bountyoperator.com to that provider.
  Source files, keys and review text are never stored.

## Quick start

### Browser

Open [bountyoperator.com](https://bountyoperator.com/) and run the built-in
example. No account, no API key.

### MCP

Run reviews from Claude Code, Codex, Cursor or any MCP client. The remote
endpoint needs nothing installed.

Claude Code:

```bash
claude mcp add --transport http bounty-operator https://bountyoperator.com/api/mcp
```

Codex:

```bash
codex mcp add bounty-operator --url https://bountyoperator.com/api/mcp
```

Cursor, in `~/.cursor/mcp.json`:

```json
{ "mcpServers": { "bounty-operator": { "url": "https://bountyoperator.com/api/mcp" } } }
```

The local server reads files by path and keeps review preparation on your
machine. It needs Node 22 or later:

```bash
npx -y https://bountyoperator.com/dl/bounty-operator-mcp.tgz
```

```bash
claude mcp add --transport stdio bounty-operator -- npx -y https://bountyoperator.com/dl/bounty-operator-mcp.tgz
codex mcp add bounty-operator -- npx -y https://bountyoperator.com/dl/bounty-operator-mcp.tgz
```

The package is coming to npm as `bounty-operator-mcp`. Until then the tarball
above is the install, and its SHA-256 is at
[bountyoperator.com/dl/SHA256SUMS.txt](https://bountyoperator.com/dl/SHA256SUMS.txt).

The remote endpoint has five tools and the local server six.
`list_profiles`, `prepare_review` and `build_packet` need no account, and
neither does the local server's `run_gauntlet_plan`. `prepare_review` takes
the three core profiles and your agent's own model writes the review. A
connection token from the account panel adds `run_review` and `account`.
`run_review` runs any profile as a hosted review and is the only way to run a
hosted one: `prepare_review` refuses it with `hosted_profile`. Token setup for
each client: [bountyoperator.com/mcp](https://bountyoperator.com/mcp) and
[mcp/README.md](mcp/README.md).

### CLI

Python 3.10 or newer. No dependencies.

```bash
pip install "git+https://github.com/bountyoperator/bounty-operator@v0.9.7"
bounty-kit agent-pack ./agent-pack.md --target "Example Protocol" --program "Example bounty"
```

Hand `agent-pack.md` to your AI agent with the scope, the known issues and the
current commit. The [CLI reference](#cli-reference) covers the other commands.

## Review profiles

Pick one profile per review. Each returns the same structure: verdict, findings
with references, counterarguments, evidence gaps, and what was checked and found
safe.

| Profile | What you get | Runs |
| --- | --- | --- |
| **Code security review** | Findings in any language, each with the path from entry point to impact in concrete values. | Core |
| **Solidity review** | Entry points, invariants, and the accounting, access-control and external-call paths that break them. | Core |
| **Challenge a draft report** | Every claim in your draft checked against the code. Claims the source does not support are listed. | Core |
| **Scope and impact fit** | The asset, the revision, the exclusions and the impact row, clause by clause. | Hosted |
| **Design intent and actors** | Who performs each step, and whether the project meant the behaviour. | Hosted |
| **Prior-art overlap** | Your finding compared with the audits and known issues you supply: same root cause, or only the same symptom. | Hosted |
| **Proof review** | Whether the proof shows the impact and not only the defect. | Hosted |
| **Severity calibration** | One level, graded on the programme's own severity table. | Hosted |
| **Triager simulation** | The three reasons a triager closes this report, ranked, with the evidence that answers each one. | Hosted |
| **Report editor** | Your report with everything a triager distrusts removed, and a list of what was cut. | Hosted |
| **Scanner triage** | Scanner output grouped by root cause into a short review queue with file and line. | Hosted |

The three core profiles, the free tools, the CLI and the MCP server are MIT and
in this repository. The gauntlet and the panel run on the hosted service, on
your own model key.

- **Core** profiles run anywhere: hosted, exported as a prompt to a chat
  subscription or a local model with paste-back, or prepared over MCP for your
  agent's own model.
- **Hosted** profiles run at bountyoperator.com or through `run_review` over
  MCP. Your files and your key go through the service to your provider, and
  the service adds the method. Free runs 1 review a day (reset at 00:00 UTC),
  any profile. The method is not in this repository: a checkout runs the hosted
  profiles on the short stand-in instructions in
  `web/src/operator-profiles.stub.mjs`.

Two runs chain the profiles on Operator:

- **Gauntlet.** One run through eight stages, in the order that ends a weak
  report early: scope, provenance, prior art, proof, severity, triager, report,
  verdict. It returns one verdict, one blocker, the cheapest action that
  removes it and a filing deadline.
- **Panel review.** Two to four models review the same files in parallel. A
  cross-examination pass keeps the findings the cited lines prove.

<a href="https://bountyoperator.com/gauntlet"><img src="docs/assets/gauntlet.png" alt="A gauntlet dossier: one verdict, the blocker, the cheapest action that removes it, the filing deadline and the eight stages" width="820"></a>

Why reports get closed, check by check:
[bountyoperator.com/method](https://bountyoperator.com/method).

## Free tools

No account. Everything runs in your browser at
[bountyoperator.com/tools](https://bountyoperator.com/tools). The CLI column is
the same check on your own machine.

| Tool | What it does | CLI |
| --- | --- | --- |
| [Report check](https://bountyoperator.com/tools/report-check) | Seventeen checks on a pasted draft: pinned commit, quoted impact row, inline proof, trusted roles, testing on a live network, AI-use disclosure, placeholders and leftover secrets. | |
| [Packet verifier](https://bountyoperator.com/tools/verify) | Recomputes the SHA-256 manifest of a review packet against your files. | |
| [Secret check](https://bountyoperator.com/tools/secret-check) | Scans a PoC or gist for keys, wallet keys and private report links before you publish it. | `bounty-kit sanitize` |
| [Acceptance rates](https://bountyoperator.com/tools/acceptance-rates) | Accepted and judged counts by bug class across 1,032 findings from 10 public Sherlock contests. | `bounty-kit pattern-stats` |
| [Slither triage queue](https://bountyoperator.com/tools/slither-focus) | Cuts `slither.json` down to 13 high-signal detectors with `file:line`. | `bounty-kit slither-focus` |

[Report templates](https://bountyoperator.com/templates) for Immunefi,
Sherlock, Cantina and HackerOne, and a Foundry PoC scaffold that ends on the
impact assertion.

## Skills and Claude Code plugin

Five skills for coding agents. In Claude Code and omp the plugin also adds the Bounty Operator MCP server.

| Skill | What it does |
|---|---|
| `challenge-report` | Checks a draft report against the code it cites and ends in one verdict |
| `solidity-review` | Solidity security review, every finding with file and line |
| `code-security-review` | Security review of any other codebase |
| `gauntlet` | Runs the eight hosted stages through the MCP server and builds the packet |
| `hunt-with-gate` | Manual only. Gates a finding before any report is written. It never submits |

Claude Code:

    claude plugin marketplace add bountyoperator/bounty-operator
    claude plugin install bounty-operator@bounty-operator

Codex, Cursor and other agents:

    npx skills add bountyoperator/bounty-operator

omp:

    omp plugin marketplace add bountyoperator/bounty-operator
    omp plugin install --scope user bounty-operator@bounty-operator

The gauntlet needs a connection token and a provider key. Set `BOUNTY_OPERATOR_TOKEN` and `BOUNTY_OPERATOR_PROVIDER_KEY` before you start the agent. Other clients add the server with the commands at [bountyoperator.com/mcp](https://bountyoperator.com/mcp).

The three review skills carry their full method and run on your own model with no account. The hosted stages run on the Bounty Operator server and are not in this repository.

On an Immunefi programme, add Immunefi Studio's own MCP server next to this one. W