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 instructions/cdump/evmole/agents-mdgit clone --depth 1 https://github.com/cdump/evmoleWhat 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 | $0.00607 | $0.00607 |
| Opus 5 | $0.00303 | $0.00303 |
| Sonnet 5 | $0.00121 | $0.00121 |
| Haiku 4.5 | $0.00061 | $0.00061 |
Grade A, and why
evmole AGENTS.md 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 yesterday.
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.
What it actually says
EVMole agent guide
EVMole extracts structured facts from deployed EVM runtime bytecode.
- This is the EVMole implementation repository. For requests to implement, debug, refactor, review, or test EVMole itself, inspect and modify the repository source. Do not use the published EVMole packages, MCP server, JSON CLI, or portable skill as a substitute for understanding the implementation. Running locally built libraries, binaries, and adapters for validation is expected.
- Treat input as deployed/runtime bytecode. Creation bytecode is not executed or stripped automatically.
- Route repository development by component:
- Core analysis algorithms and the public Rust API:
src/andCargo.toml. - Python bindings:
src/interface_py.rsandpython/. - JavaScript/WASM bindings and the JSON CLI:
src/interface_js.rs,src/interface_wasm.rs, andjavascript/. - Go bindings:
go/. - MCP server, shared agent adapter, portable skill, schemas, and agent tests:
javascript/src/agent_api.mjsandagent/.
- Core analysis algorithms and the public Rust API:
- Use
agent/skills/evm-bytecode-analysis/SKILL.mdonly when the user asks to analyze supplied deployed runtime bytecode, or when changing or testing the CLI, MCP server, portable skill, schemas, interpretation behavior, or agent integration. - When helping a separate project integrate EVMole, use the existing binding
for the project's language: the
evmoleRust crate,evmoleJavaScript package,evmolePython package, orgithub.com/cdump/evmole/go. Start withREADME.mdand the language-specific README. - Reuse the shared agent adapter when changing the CLI, MCP server, skill, or agent tests; do not duplicate its request, response, validation, or pagination logic.
- Validate the components changed:
- Rust:
cargo fmt --check,cargo test, andcargo clippy --all-features -- -D warnings. - JavaScript and the agent adapter: build with
npm --prefix javascript run build, then run focused tests withnpm --prefix javascript testandnpm --prefix agent/mcp test. - Go: build the embedded WASM with
make -C go wasm, then runmake -C go test. - Python: build with Maturin and run
python3 python/test_python.py; use the release workflow for the supported Python and platform matrix.
- Rust:
- When reporting bytecode-analysis results, describe inferred arguments, mutability, and storage information as inferred, not verified source-level facts.
- Preserve unrelated benchmark providers, datasets, result directories, and user changes.
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.
- yesterday First seen · 47 lines · 607 tokens per session scan A 8c40f0975d86
evmole AGENTS.md is an instructions file published in the GitHub repository cdump/evmole (461 stars, last pushed 17d ago), licensed MIT. It adds 607 tokens to every session, about $0.0030 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 2026-08-30.
Other instructions, from other repositories
prb-math AGENTS.md
Instructions for PaulRBerg/prb-math, covering prbmath, stack, structure, commands and development.
solidity-language-server AGENTS.md
Instructions for asyncswap/solidity-language-server, covering agents, prerequisites: ast identity model, the two id types, the src string format and why file ids are unstable from solc.
mcp-tenderly CLAUDE.md
Claude Code instructions for py-zoid/mcp-tenderly, covering mcp-tenderly, commands, layout, constraints worth knowing before you change anything and testing against real tenderly.
prb-math CLAUDE.md
Instructions for PaulRBerg/prb-math, a project described as: Solidity library for advanced fixed-point math.
tevm CLAUDE.md
Claude Code instructions for evmts/tevm, covering tevm monorepo commands and style guide, project overview, forking implementation, json-rpc support and commands.
safe_wallet AGENTS.md
Instructions for ascenx/safe_wallet, covering agents.md, 1. 沟通原则, 2. 工作方式, 3. 代码修改边界 and 4. git 规则.