Skip to content

Commit c9057d2

Browse files
author
lloyd-c137
committed
feat: Autonomous Bounty-Hunting Agent — proper self-contained implementation
Complete rewrite: the agent is now fully self-contained with zero external system dependencies. No OpenClaw, no Semgrep, no Python — just Node.js + npm. Key changes: - Self-contained agent with Anthropic Claude SDK for AI orchestration - Octokit for all GitHub operations (discover, audit, submit) - SKILL.md for Claude Code integration - Zero system dependencies — works anywhere with Node.js 18+ - 11 security patterns in pure regex (no Semgrep needed) - AI fix generation with Claude (Anthropic) + SiliconFlow fallback Closes #861
1 parent 0bb39b1 commit c9057d2

6 files changed

Lines changed: 433 additions & 0 deletions

File tree

agents/spike/.env.example

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
GITHUB_TOKEN=ghp_your_token_here
2+
ANTHROPIC_API_KEY=sk-ant_your_key_here
3+
SILICONFLOW_KEY=sk-your_key_here

agents/spike/README.md

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,7 @@
1+
# Spike 🎯
2+
3+
Autonomous Bounty-Hunting Agent for SolFoundry.
4+
5+
```bash
6+
npm install && spike pipeline
7+
```

agents/spike/SKILL.md

Lines changed: 32 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,32 @@
1+
# Spike — Autonomous Bounty-Hunting Agent
2+
3+
> "You're gonna carry that weight." — Spike Spiegel
4+
5+
## Description
6+
An autonomous AI agent that discovers open-source security bounties, audits repositories for vulnerabilities, generates AI-powered fixes, and submits PRs. Built for SolFoundry's bounty ecosystem.
7+
8+
## Commands
9+
- `/spike discover` — Scan Algora, GitHub issues, and Security Advisories for bounty opportunities
10+
- `/spike scan <owner/repo>` — Perform deep security audit on a repository
11+
- `/spike pipeline` — End-to-end: discover → audit → generate fixes → report
12+
13+
## Requirements
14+
- Node.js 18+
15+
- GITHUB_TOKEN (GitHub API)
16+
- ANTHROPIC_API_KEY (for AI fix generation via Claude)
17+
18+
## Architecture
19+
Four-agent orchestration:
20+
1. **Discovery Agent** — Finds bounty opportunities across platforms
21+
2. **Audit Agent** — Static analysis with 11 security patterns (zero dependencies)
22+
3. **Fix Agent** — AI-powered fix generation via Anthropic Claude + SiliconFlow fallback
23+
4. **Submit Agent** — GitHub API integration for PR submission
24+
25+
## Example
26+
```bash
27+
export GITHUB_TOKEN=ghp_...
28+
export ANTHROPIC_API_KEY=sk-ant_...
29+
npx spike discover
30+
npx spike scan expressjs/express
31+
npx spike pipeline
32+
```

agents/spike/bin/spike

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,2 @@
1+
#!/usr/bin/env node
2+
require('../src/agent.js');

agents/spike/package.json

Lines changed: 20 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,20 @@
1+
{
2+
"name": "spike",
3+
"version": "1.0.0",
4+
"description": "Autonomous bounty-hunting agent",
5+
"main": "src/agent.js",
6+
"scripts": {
7+
"test": "echo \"Error: no test specified\" && exit 1"
8+
},
9+
"keywords": [],
10+
"author": "",
11+
"license": "ISC",
12+
"dependencies": {
13+
"@anthropic-ai/sdk": "^0.95.2",
14+
"@octokit/rest": "^22.0.1",
15+
"dotenv": "^17.4.2"
16+
},
17+
"bin": {
18+
"spike": "./bin/spike"
19+
}
20+
}

0 commit comments

Comments
 (0)