lfglabs-dev/verity/public
Claude Code and agent skill by lfglabs-dev/verity
Trust tier 2
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Install
npx vskill@latest install lfglabs-dev/verity/publicSKILL.md
Verity Agent Skill
Use this skill when working in the Verity repository or docs. Verity is a Lean 4 framework for writing smart contracts that compile to EVM bytecode with machine-checked correctness proofs.
First Principles
- Treat Lean files in
Contracts/<Name>/as the source of truth. - Keep specs, EDSL implementation, generated artifacts, and proofs aligned.
- Do not use
sorry,admit, newaxioms, or unchecked assumptions to make proofs pass. - Prefer existing Verity primitives and proof patterns before adding abstractions.
- Read
/llms.txtfirst for the compact working model, then fetch/llms-full.txtwhen broader context is needed.
Add A Contract
- Scaffold with
python3 scripts/generate_contract.py <Name> --fields ... --functions ...when the contract shape fits the generator. - Implement storage, specs, and EDSL functions under
Contracts/<Name>/. - Register compiler integration following the existing contract examples.
- Prove each public operation with the
_meets_specconvention. - Run
lake build. - If compiler behavior changes, run the relevant Foundry and property tests.
Primary docs:
/first-contract.md/guides/add-contract.md/edsl.md(EDSL reference hub; subpages/edsl/storage.md,/edsl/computation.md,/edsl/control-flow.md,/edsl/functions.md,/edsl/external-calls.md)/proof-techniques.md/guides/debugging-proofs.md
Port Solidity To Verity
- Map Solidity state to Verity storage spaces before writing code.
- Translate modifiers into explicit guard predicates and
requirechecks. - Model external calls, precompiles, or oracle inputs as explicit trust-boundary assumptions unless the repo already has a verified surface.
- Use
/guides/production-solidity-patterns.mdfor ERC-7201 namespaces, proxy patterns, ECMs, events, and low-level mechanics. - Validate emitted trust, assumption, layout, and layout-compat reports.
Primary docs:
/guides/solidity-to-verity.md/guides/production-solidity-patterns.md/compiler.md/trust-model.md
Prove Specs
- The default theorem shape is
<operation>_meets_spec. - Most storage-only operations close by unfolding the operation, slot names, and spec, then using
simp. - For guard-protected operations, unfold
bind,Contract.run, andContractResult.snd, then pass the guard hypothesis tosimp. - For list or aggregate invariants, isolate helper lemmas near the proof instead of growing tactic scripts blindly.
- When stuck, inspect generated goals before changing the spec. A failing proof often means the spec, state shape, or revert path is underspecified.
Audit Trust
lake buildverifies Lean proofs.- Use compiler report flags for trust-sensitive changes:
--trust-report,--assumption-report,--layout-report, and--layout-compat-report. - Use deny flags to fail closed in CI or release builds.
--deny-unsafeis the broad catch-all. - Cross-check trust reports against
AXIOMS.mdand/trust-model.md.
Useful Agent URLs
/llms.txt: compact agent index./llms-full.txt: all docs concatenated as Markdown./api/docs/_index: JSON documentation index./api/docs/_all: all docs concatenated through the API./<page>.md: raw Markdown for any docs page./.well-known/llms.txt: discovery alias./.well-known/llms-full.txt: discovery alias./.well-known/skill.md: discovery alias./.well-known/agent-skills: skill discovery index.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingdocs-site/public/skill.md
Popularity
- Repository stars
- 144
- Repository forks
- 18
- 7-day trend score
- 25
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/lfglabs-dev/verity/public)