wallet-desk-brief
Skill2 starsChanged 41 days agoApache-2.0View on GitHub
Use this skill when the user supplies one explicit non-zero EVM address and wants a bounded, read-only wallet attribution-and-holdings snapshot—for example, 'what public label is associated with this wallet?', 'show this address's token footprint on the supported networks', or 'cross-check an Arkham label against holdings context'. It runs a fixed pair: Alchemy token holdings across five named networks and Arkham's probabilistic address intelligence. Require a free live quote and explicit cost approval before paying. It does not prove real-world ownership, cover every EVM chain, provide financial advice, or execute trades.
In this repository19 other files it gives its agents.
- account-intelSkill
- comprehensive-enrichmentSkill
- email-campaignSkill
- enrich-waterfallSkill
- financial-intelSkill
- find-twitter-influencersSkill
- gtm-enrichment-deepSkill
- gtm-enrichment-smartSkill
- lead-enrichmentSkill
- perplexity-searchSkill
- person-lookupSkill
- recent-funding-roundsSkill
- sales-prospectingSkill
- scrapecreatorsSkill
- self-evolving-agentSkill
- social-intelSkill
- stock-direction-signalsSkill
- twitter-researchSkill
- vc-ai-infra-scoutSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 6 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
Use this skill for public-blockchain research on one complete address when the user wants both:
- Distinctive: line 28
Do not use it for an asset/ticker outlook, protocol research, transaction-flow analysis, compliance decisions, Solana-only addresses, ENS …
- Very specific: line 36
Do not accept 0x..., an ENS name, a transaction hash, or a sample fallback.
- Distinctive: line 66
Run the free live gate:
- Very specific: line 79
If wallet setup or funding is missing, treat it as a separate action requiring separate approval.
- Distinctive: line 81
After approval, arm only the approved cumulative session cap, execute exactly once, and disarm immediately after success or failure:
- Distinctive: line 133
Do not calculate a complete net worth unless every included holding has a valid, time-stamped price and the result is demonstrably complete.
- Distinctive: line 144
The network scope is fixed and incomplete: No holdings on other EVM chains, Solana, Bitcoin, or exchange-internal accounts are retrieved.
- Very specific: line 146
Polygon's Portfolio API identifier is
matic-mainnet: Do not substitute a node-RPC hostname label in this request body.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
wallet-desk-brief
A fixed two-call, read-only snapshot for one explicit EVM address. It retrieves token holdings across five named networks and a separate probabilistic address label, then returns two raw responses for the calling agent to reconcile. The manifest does not prove ownership, merge results, score risk, or place trades.
When To Use
Use this skill for public-blockchain research on one complete address when the user wants both:
- a bounded token-holdings snapshot; and
- any label or entity association the attribution provider returns.
Do not use it for an asset/ticker outlook, protocol research, transaction-flow analysis, compliance decisions, Solana-only addresses, ENS resolution, private identity discovery, or order execution. A request to buy, sell, transfer, or freeze assets is outside this read-only workflow.
Input Gate
Require address to match ^0x[0-9a-fA-F]{40}$ and reject the all-zero address.
Do not accept 0x..., an ENS name, a transaction hash, or a sample fallback.
Echo the normalized address and fixed network scope before verification so the
user can catch a mistargeted query.
Public-chain visibility does not make an attribution infallible. Do not infer a private person's identity, intent, criminality, solvency, or control of the address from a label or token balance alone.
Rails And Fixed Pipeline
On 2026-10-04 the free probe of both exact manifest POSTs reported
mode: routed-x402 through the SELAT Router: Alchemy at $0.001 (cap $0.002)
and Arkham at $0.21 (cap $0.30). The expected total is $0.211, and the
per-step caps sum to $0.302. Alchemy's 402 also lists a
GatewayWalletBatched offer, so report the mode that verify prints instead of
assuming one:
- Alchemy token holdings — one POST request across Ethereum, Base, Polygon PoS, Arbitrum, and Optimism mainnets. Metadata, available USD prices, native tokens, and ERC-20 tokens are requested.
- Arkham address intelligence — one POST request with the same address. The optional chain field is omitted, so preserve the chain returned by the provider and do not generalize the label to every chain.
This is a fixed pipeline, not a menu or fallback. The runner executes both calls in order and may continue after an individual failure.
Workflow
- Validate the explicit address locally using the Input Gate above.
Run the free live gate:
selat skill validate ./skills/wallet-desk-brief selat skill verify ./skills/wallet-desk-brief \ --address <explicit-non-zero-evm-address> \ --live-probe- Show both live routes and quotes, the expected cumulative cost, the sum of source-defined per-call caps, the proposed absolute session cap, and current transactability cautions. Explain that a paid application error may still be charged. Stop for explicit approval.
- If wallet setup or funding is missing, treat it as a separate action requiring
separate approval. Do not run
selat initorselat fundimplicitly. After approval, arm only the approved cumulative session cap, execute exactly once, and disarm immediately after success or failure:
selat budget start --amount <approved-session-cap> selat skill run wallet-desk-brief \ --address <the-exact-approved-evm-address> selat budget stop- Inspect per-step status and payment history before any retry. Refresh the affected quote and obtain new approval; never rerun the whole pair automatically when only one step failed.
Inputs And Outputs
| Param | Required | Default | Description |
address | yes | none | One explicit non-zero EVM address used unchanged by both reads. |
The runner returns two independent raw JSON responses. The calling agent—not the manifest—must normalize and synthesize them.
For holdings, preserve:
- the queried address and exact network for every token;
- raw balance, decimals, symbol/name, and token contract when available;
- price value and price timestamp when supplied;
- top-level
partialErrorsand per-token errors; - pagination or truncation indicators.
For attribution, preserve:
- returned chain, entity, label, and contract/user-address flags;
- null or missing attribution as unlabeled, not anonymous or safe;
- provider confidence or evidence fields when present.
Synthesis Rules
Return a compact, non-advisory brief containing:
- Exact queried address, retrieval time, and the five-network holdings scope.
- Arkham's returned label/entity and chain, explicitly described as a probabilistic provider attribution—not verified ownership.
- Holdings by network, with raw balance conversions and price timestamps only when the response supplies enough metadata.
- Partial-network failures, missing prices, pagination limits, spam/airdropped token caveats, and other evidence gaps.
- Agreements or conflicts between the label and holdings without treating a token balance as proof of identity, intent, endorsement, or control.
- Per-step success/failure, any charged error, and final settled cost.
Do not calculate a complete net worth unless every included holding has a valid, time-stamped price and the result is demonstrably complete. Do not treat receipt of a spam token or dust as evidence that the wallet chose or endorsed it. Keep provider endpoint URLs and raw JSON out of the user-facing brief while citing appropriate public evidence where available.
Gotchas
- The documented Alchemy contract is POST. The body contains an
addressesarray and explicit networks. A GET/query request can expose a payment challenge but does not prove the post-payment business request is valid. - The network scope is fixed and incomplete. No holdings on other EVM chains, Solana, Bitcoin, or exchange-internal accounts are retrieved.
- Polygon's Portfolio API identifier is
matic-mainnet. Do not substitute a node-RPC hostname label in this request body. - Attribution is not ownership proof. Labels can be incomplete, contested, or revised as intelligence changes.
- A 200 can contain partial errors. Inspect Alchemy's top-level network errors and individual token errors before calling the snapshot complete.
- Caps are not prices. Re-quote before each paid run and use the step-cap sum
(
$0.302) as the absolute maximum session exposure. The top-levelmaxAmountis only a per-step fallback, not a run cap. - Alchemy delivery is weak. The live quote's transactability extension reports 7% all-time delivery over 15 captured payments, mostly upstream 5xx after capture. Say so before approval, and rehearse one paid Alchemy call before any live demo. Arkham's 100% comes from one sample.
- Paid failures may charge. The runner may continue. Never auto-retry.
- No default address. Missing or malformed input must fail before any probe.
Validation
- Static:
selat skill validate ./skills/wallet-desk-brief - Missing-input gate:
selat skill verify ./skills/wallet-desk-brief --live-probemust fail before probing either endpoint. - Free live gate:
selat skill verify ./skills/wallet-desk-brief --address 0x28C6c06298d514Db089934071355E5743bf21d60 --live-probe - Single-step free probes: see
references/endpoints.md. - Paid confirmation: add
--payonly after a fresh quote, explicit approval, and an armed session budget. Never reuse approval after an address, route, quote, or cap changes.
References
manifest.json— the machine-readable fixed payment recipe.references/endpoints.md— request contracts, current routes, quotes, and interpretation limits.../../references/agent-skill-authoring-sop.md— authoring standard.- Alchemy Tokens By Wallet documentation — https://www.alchemy.com/docs/data/portfolio-apis/portfolio-api-endpoints/portfolio-api-endpoints/get-tokens-by-address
- Arkham API Guide — https://arkm.com/docs
Alchemy and Arkham are third-party services. Their names and trademarks belong to their respective owners and are used only to identify the endpoints called.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

