Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/omermaksutii/rugproof/exploit-livegit clone --depth 1 https://github.com/omermaksutii/RugProofWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00023 | $0.00646 |
| Opus 5 | $0.00012 | $0.00323 |
| Sonnet 5 | $0.00005 | $0.00129 |
| Haiku 4.5 | $0.00002 | $0.00065 |
Grade A, and why
exploit-live scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/exploit-live — exploit a deployed contract on a fork
For when the user has a finding from /audit-live and wants to verify it works on the real deployment.
Procedure
Step 1 — Fork the chain
mcp__anvil__fork(chain=<chain>, block=latest)
Capture the RPC URL.
Step 2 — Pull source if needed
If we don't already have the source from /audit-live, pull it via block-explorer.
Step 3 — Write the PoC against the fork
Same template as /exploit but using vm.createFork and vm.activeFork to target the live deployment:
function setUp() public {
forkId = vm.createFork(ANVIL_FORK_URL);
vm.selectFork(forkId);
target = Vault(payable(LIVE_ADDRESS));
}
function test_Exploit_Live() public {
// attempt against the *actual deployed state*
}
Step 4 — Run
Use forge-runner MCP with --fork-url pointed at the anvil instance.
Step 5 — Output
If the test passes:
✗ EXPLOITABLE on live deployment
Chain: berachain
Block: 12345678
Contract: 0xabc...
Finding: REENT-001
Value extractable: ~$420K (at current price)
PoC: test/live-exploits/ExploitLive-REENT-001.t.sol
If the test fails (i.e. the exploit doesn't work in practice):
✓ Finding REENT-001 does NOT exploit the live deployment as written.
Possible reasons: live state differs from setUp assumption, finding requires
preconditions not met by current state, or finding is a false positive.
This is a valuable output — false positives caught.
Responsible disclosure
If the exploit works against a contract not owned by the user, the output should include a prominent disclosure reminder:
⚠ RESPONSIBLE DISCLOSURE
This exploit works against a live deployment that you may not own.
DO NOT execute on mainnet. Coordinate disclosure via:
- Immunefi: https://immunefi.com
- Protocol's bounty: (check the protocol's docs)
- SEAL 911: https://seal-911.com (emergency)
Notes
- Always fork at a recent block to reflect current state.
- Don't share the PoC publicly until the protocol has patched. Use
/bountyto generate a disclosure-ready report and/bounty-submitto package it for Immunefi.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 88 lines · 23 tokens per session scan A f8738cd72a98
exploit-live is a command published in the GitHub repository omermaksutii/RugProof (9 stars, last pushed 1mo ago), licensed MIT. It adds 23 tokens to every session and 646 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
krait-fuzz
Run an invariant-based fuzzing campaign: Understand → Extract Invariants → Generate Foundry Tests → Run & Fix Iteratively → Report.
krait-poc
Write and run a valid Foundry proof-of-concept that proves (or disproves) a Solidity exploit by asserting the actual harm on a forked chain or against local source.
krait-quick
Run a streamlined audit: Recon → Detection → Verification → Report. Skips state inconsistency analysis and cross-feed iteration for speed.
fix-btc-e2e-p11
Fix errors in BTC E2E test (Pattern 11: P2TR Tapscript M-of-N).
fix-btc-e2e-p6
Fix errors in BTC E2E test (Pattern 6: P2WSH 2-of-3 Multisig).
phase6-start
Command "phase6-start" from kota1026/quantum-shield, covering phase 6 $arguments 開始, step 1: 進捗状況を確認, step 2: 未完了画面を特定, step 3: 必須ファイルを読み込む and step 4: インフラ確認.