uniswap/uniswap-ai/permissioned-pools-configurator
Claude Code and agent skill by uniswap/uniswap-ai
- community
Checking that the SKILL.md file is still in the repository.
Interactively collect and validate the parameters for a Uniswap v4 Permissioned Pool setup - chain, underlying token, allowlist checker, adapter owner, verification deposit, the four wrapper registrations, hook, and pool pricing - and emit the JSON config the deployer skill consumes. Use when the user says "configure a permissioned pool", "permissioned pool config", "adapter parameters", "permissioned pool setup parameters", "PermissionsAdapter configuration", or asks to set up parameters for a permissioned pool before deploying. Covers contract mechanics only - does not constitute legal, financial, investment, tax, or compliance advice, and is not a compliance review.
Install
npx vskill@latest install uniswap/uniswap-ai/permissioned-pools-configuratorSKILL.md
Permissioned Pools Configurator
Interactive bulk-form configurator for a Uniswap v4 Permissioned Pool. Collects every parameter the setup journey needs, validates each one, and displays a single JSON configuration object that the permissioned-pools-deployer skill consumes to run the on-chain sequence. For the contract mechanics behind any of these parameters, see `permissioned-pools-issuer`.
Runtime Compatibility: This skill usesAskUserQuestionto collect parameters inbatches. IfAskUserQuestionis not available in your runtime, collect the sameparameters through natural language conversation instead, in the same batch groupingsand with the same validation after each group.
Scope and Disclaimer
This is a configuration tool, not deployment guidance and not a compliance review.
- ✅ It collects and validates the inputs a permissioned-pool setup needs and produces a
JSON document. It does not call any contract, does not broadcast anything, and holds no signing keys.
- ✅ Treat this as contract mechanics only. This skill does not constitute legal, financial, investment, or tax advice, and it is not a compliance review of your token, your allowlist, your KYC or AML program, or your configuration. It covers what each parameter means and what breaks if it is wrong.
- ✅ It never invents a contract address. Addresses you supply are recorded as given;
addresses you do not yet have — whether because they still need to be looked up against the deploy guide's table or because the issuer has not decided on them yet — are stored as the literal string "RESOLVE", never guessed or recalled from memory.
- ✅ Review the emitted configuration yourself, and re-resolve every address it marks for
resolution, before handing it to a deployment flow.
- ✅ Read the repo's usage guidelines. The repo root
DISCLAIMER.mdgoverns every skill in
this repository: they are provided as is without warranty, they do not constitute legal, financial, investment, or tax advice, and it sets out use limits plus an AI-disclosure duty that applies when you use a skill to generate financial information and present it directly to individuals or consumers. Point the user to it.
State this framing inline (quoted or in your own words) and continue with the configuration flow in the same response — this is reference and data-collection content, not an action-oriented deployment step, so no acknowledgment gate is needed here. The deployer skill gates on acknowledgment before anything gas-spending.
What This Config Is (and Isn't)
The output is a pre-deployment plan, not a record of what already exists on-chain.
- The `PermissionsAdapter` address does not appear in this config. It does not exist
until the deployer's Step 2 (createPermissionsAdapter) creates it. If the issuer already has a deployed adapter for this token, that is a re-run/import scenario this skill does not currently handle — collect its address as a note and confirm with the deployer skill's preflight checks instead of adding it here. This declines only the adapter address itself — it does not block the rest of the request. Continue validating and including every other field the user supplied (or marked "RESOLVE") in the emitted configuration exactly as the flow below describes.
- The underlying permissioned token is the only token address this config records.
See the two-addresses distinction in permissioned-pools-issuer — the pool currency the deployer builds later is the adapter, derived at deploy time, not configured here.
- Wrapper and hook addresses are chain-specific and change over time. This skill asks
whether you already have them resolved for your target chain; if not, it marks them for resolution rather than filling in a value.
Configuration Flow
Collect parameters in four batches of at most four questions each. After each batch, validate every answer against `config-schema.md` before moving on, and show a running summary of what has been collected and what remains. Field-by-field meaning, how to choose each value, and what breaks if it is wrong are all in `parameter-reference.md`.
Front-loaded or skip-ahead answers
A user may answer several batches' worth of questions in one free-form message, or ask to skip the remaining questions and finish immediately. Do not withhold the whole configuration while asking whether to proceed. Instead: validate everything the user did supply against config-schema.md, apply the literal string "RESOLVE" (or null for allowlistChecker.address when the checker is to-be-deployed) to every address-type field left open, and produce that partial configuration in the same response. Fields with no "RESOLVE" sentinel — the verification deposit amount, pool currency, fee tier, tick spacing, starting price, and seeding — have no safe default; flag each one explicitly as still needing an answer rather than guessing at it, and ask only about those. This includes the verification deposit amount even though Batch 2 lists "1 wei" as the recommended option: a menu option a user has not actually picked is not a value the skill may silently apply on their behalf — treat an unanswered verification deposit amount exactly like the other sentinel-less fields, never as an already-resolved "1". Skipping the batch-by-batch confirmation step never means skipping validation — every rule above still applies to whatever the user did supply.
Batch 1: Network, Token & Ownership (4 questions)
Question 1 — Network
- Prompt: "Which network is this permissioned pool for?"
- Options: "Ethereum Mainnet (chain ID 1)", "Unichain (chain ID 130)", "Sepolia (chain ID
11155111)", Custom chain ID (via "Other")
- Note in the prompt: availability of the permissioned-pools contracts on a given chain is
not something this skill asserts — confirm the chain has a row in the deploy guide's #deployment-addresses table before proceeding.
- Store:
chainId
Question 2 — Underlying permissioned token
- Prompt: "What is the address of the permissioned ERC-20 being listed?"
- Options: "Not deployed yet" (stores the literal string
"RESOLVE"), custom address
(via "Other")
- Validation:
^0x[a-fA-F0-9]{40}$, or the literal string"RESOLVE"if not yet deployed - Store:
permissionedToken
Question 3 — Allowlist checker
- Prompt: "Do you already have an allowlist checker deployed for this token?"
- Options, each mapped to the exact
allowlistCheckervalue it produces: - "Yes, use an existing address" (then ask for the address via "Other") →
mode: "existing", address set to that address.
- "No, I still need to deploy one" →
mode: "to-be-deployed",addressset tonull. - Validation: address form above when "existing" is chosen
- Store:
allowlistChecker.mode(existing|to-be-deployed),allowlistChecker.address
Question 4 — Adapter owner
- Prompt: "What address should own the
PermissionsAdapter(initialOwner)?" - Options: "Not yet decided — mark for resolution" (stores the literal string
"RESOLVE"), custom address (via "Other")
- Note in the prompt: this key can change the allowlist checker, add or remove wrappers,
pause and resume swapping, and force-exit any LP position — the published guide recommends a multisig. Do not pick this in a hurry just to finish the flow; use the resolution marker instead if the multisig or key is not decided yet.
- Validation:
^0x[a-fA-F0-9]{40}$and not the zero address, or the literal string
"RESOLVE"
- Store:
adapterOwner
Validate: confirm chain ID is a positive integer, the token address is either a valid address or the literal string "RESOLVE", the checker address (if "existing" was chosen) matches the address regex, and the adapter owner is either a valid non-zero address or the literal string "RESOLVE". Show a summary of the four collected values, flagging any field still marked "RESOLVE".
Batch 2: Verification & Core Wrappers (4 questions)
Question 1 — Verification deposit amount
- Prompt: "How much of the underlying token should be deposited for verification?"
- Options: "1 wei (recommended)", Custom amount (via "Other")
- Note in the prompt: this deposit becomes the adapter's mintable headroom, there is no
withdraw function, and 1 wei is sufficient — see Parameter Reference before choosing anything larger.
- "Recommended" describes the value if asked and picked — it is never a license to fill
in "1" on the user's behalf when this question was skipped. See Front-loaded or skip-ahead answers above.
- Store:
verificationDepositAmount
Question 2 — `PermissionedPositionManager` address
- Prompt: "Do you have the
PermissionedPositionManageraddress resolved for this
chain?"
- Options: "Not yet resolved — mark for resolution" (stores the literal string
"RESOLVE"), custom address (via "Other")
- Validation: address form, or the literal string
"RESOLVE" - Store:
allowedWrappers.permissionedPositionManager
Question 3 — Universal Router (v2.2+) address
- Prompt: "Do you have the Universal Router
#v2.2(or higher) address resolved for
this chain?"
- Options: "Not yet resolved — mark for resolution" (stores the literal string
"RESOLVE"), custom address (via "Other")
- Note in the prompt: the plain
UniversalRouterdeployments key is a different,
non-permissioned router — only the #v2.2 deployment works here.
- Validation: address form, or the literal string
"RESOLVE" - Store:
allowedWrappers.universalRouterV2_2
Question 4 — `PermissionedHooks` address
- Prompt: "Do you have the
PermissionedHooksaddress resolved for this chain, for
setAllowedHook?"
- Options: "Not yet resolved — mark for resolution" (stores the literal string
"RESOLVE"), custom address (via "Other")
- Note in the prompt: this is the shared hook address, registered with
setAllowedHook on the position manager — it is never registered as a wrapper.
- Validation: address form, or the literal string
"RESOLVE" - Store:
hook
Validate: confirm the deposit amount is a positive integer (not zero) expressed in the token's smallest unit, and that every address field is either a valid address or the literal string "RESOLVE" — never a guessed or invented value. Show a summary.
Batch 3: Remaining Wrappers & Pool Currency (4 questions)
Question 1 — `V4Quoter` address
- Prompt: "Do you have the
V4Quoteraddress resolved for this chain?" - Options: "Not yet resolved — mark for resolution" (stores the literal string
"RESOLVE"), custom address (via "Other")
- Validation: address form, or the literal string
"RESOLVE" - Store:
allowedWrappers.v4Quoter
Question 2 — `MixedRouteQuoterV2` address
- Prompt: "Do you have the
MixedRouteQuoterV2address resolved for this chain?" - Options: "Not yet resolved — mark for resolution" (stores the literal string
"RESOLVE"), custom address (via "Other")
- Validation: address form, or the literal string
"RESOLVE" - Store:
allowedWrappers.mixedRouteQuoterV2
Question 3 — Paired currency
This is the start of the file. Read the full SKILL.md on GitHub.
Security
- Security scanPassed, 93 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 repositoryCheckingpackages/plugins/uniswap-permissioned-pools/skills/permissioned-pools-configurator/SKILL.md
Popularity
- Repository stars
- 229
- Repository forks
- 35
- 7-day trend score
- 27
Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 4 Sep 2026, a month ago.
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/uniswap/uniswap-ai/permissioned-pools-configurator)