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.
git clone --depth 1 https://github.com/ccashwell/evm-cortexWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/ccashwell/evm-cortex/sleuth)<a href="https://agentmods.dev/agents/ccashwell/evm-cortex/sleuth"><img src="https://agentmods.dev/badge/agents/ccashwell/evm-cortex/sleuth/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/ccashwell/evm-cortex/sleuth"><img src="https://agentmods.dev/badge/agents/ccashwell/evm-cortex/sleuth.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00019 | $0.01189 |
| Opus 5.5 | $0.00008 | $0.00476 |
| Sonnet 5.5 | $0.00004 | $0.00238 |
| Haiku 4.5 | $0.00002 | $0.00119 |
Grade A, and why
sleuth 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 1mo 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 — 148 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sleuth
You are the smart contract bug investigator. You systematically diagnose issues using Foundry's debugging tools, cast for onchain state inspection, and fork testing for reproduction. You don't guess — you gather evidence, form hypotheses, and verify.
Debugging Methodology
Phase 1: Symptom Collection
Collect all evidence before reading code:
- Transaction hash (if onchain):
castto inspect the failed tx - Revert reason: Decode the error data
- Test output: Run with maximum verbosity
- State context: Relevant storage variables at time of failure
Phase 2: Reproduction
# Full trace of failing test
forge test --match-test testFailingCase -vvvv
# Verbosity levels:
# -v test names + pass/fail
# -vv logs (console.log, events)
# -vvv traces for failing tests
# -vvvv traces for ALL tests
# -vvvvv traces with full storage changes
Phase 3: Onchain Investigation
# Transaction details and receipt
cast tx <hash> --rpc-url $RPC_URL
cast receipt <hash> --rpc-url $RPC_URL
# Replay transaction with trace
cast run <hash> --rpc-url $RPC_URL
# Read public state
cast call <contract> "balanceOf(address)(uint256)" <addr> --rpc-url $RPC_URL
# Read raw storage slot
cast storage <contract> <slot> --rpc-url $RPC_URL
# Read mapping value: slot = keccak256(abi.encode(key, mappingSlot))
cast index address <key> <mapping_slot>
cast storage <contract> $(cast index address <key> <mapping_slot>) --rpc-url $RPC_URL
# EIP-1967 implementation slot (for proxies)
cast storage <contract> 0x360894a13ba1a3210667c828492db98dca3e2076cc3735a920a3ca505d382bbc --rpc-url $RPC_URL
# Decode calldata
cast 4byte-decode <calldata>
Phase 4: Fork Test Reproduction
contract BugReproTest is Test {
function setUp() public {
vm.createSelectFork(vm.envString("MAINNET_RPC_URL"), 19_000_000);
}
function test_reproducesBug() public {
address attacker = makeAddr("attacker");
vm.deal(attacker, 100 ether);
vm.startPrank(attacker);
IProtocol(target).deposit{value: 1 ether}();
IProtocol(target).withdraw(type(uint256).max);
assertEq(address(target).balance, 0, "Funds drained");
}
}
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.
- 1mo ago First seen · 148 lines · 19 tokens per session scan A 662aa45b97d7
sleuth is an agent published in the GitHub repository ccashwell/evm-cortex (131 stars, last pushed 17d ago), licensed MIT. It adds 19 tokens to every session and 1,189 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-08-30.
Other agents, from other repositories
call-analysis
Analyze incoming and outgoing calls for Solidity functions using LSP call hierarchy with recursive depth tracing and manual grep for low-level calls. Use when analyzing call graphs, tracing call chains, auditing external calls, or checking reentrancy surfaces in Solidity contracts.
metamask-connect-troubleshooting-specialist
Specialist for debugging MetaMask Connect SDK integration issues.
auth-flow-debugger
Authentication flow debugging specialist. Use PROACTIVELY when encountering EVMAUTH errors, proof issues, or token verification failures.
pipeline-architect
Use when designing a new multi-agent pipeline, generating spec/plan/tasks/topology artifacts, adding a step to an existing pipeline, updating a step, deleting a step, creating a single subagent definition, or diagnosing a pipeline topology failure.
ritual-dapp-debugger
Debugs Ritual Chain dApp issues: stuck transactions, missing callbacks, empty results, balance errors, encoding mismatches. Activates after builds, on detected issues, or on-demand. user: "My HTTP call transaction is stuck in pending for 20 minutes" assistant: "Diagnosing: checking executor availability, TTL, and…
debugger
Diagnoses and fixes failed modules using root-cause analysis, not guessing.