What makes Get with Receipt notable?

1K downloads. Trust evidence available

How should Get with Receipt be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Get with Receipt

Name an outcome—from web search to data lookup. Receipt finds eligible paid tools, shows the seller and price, clears agent purchasing under spending limits and purchase approval, and returns the result with signed proof.

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.0k

Updated

Oct 11, 2026

Version

1.0.7

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1K downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 1K downloadsadoption · observed Oct 11, 2026
Latest release: 1.0.7release · observed Aug 2, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s17cshqyqnt6m7s2k9f0tnvrwh8bafxz:get-with-receipt

Install using `clawhub skill install s17cshqyqnt6m7s2k9f0tnvrwh8bafxz:get-with-receipt` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/receiptprotocol/get-with-receipt before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-receiptprotocol-get-with-receipt/snapshot"

Run-check

$0.02 USD

1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.

Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.

Documentation

CLAWHUB

146,065 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: get-with-receipt
description: Name an outcome—from web search to data lookup. Receipt finds eligible paid tools, shows the seller and price, clears agent purchasing under spending limits and purchase approval, and returns the result with signed proof.
homepage: https://receiptprotocol.com/docs/openclaw
user-invocable: true
metadata:
  {
    "openclaw":
      {
        "emoji": "🧾",
        "homepage": "https://receiptprotocol.com/docs/openclaw",
        "version": "1.0.7",
        "publisher": "@receiptprotocol",
        "category": "Finance",
      },
  }
---

# Get with Receipt (v1.0.7)

Tell your agent what you need. Receipt finds an eligible way to get it done, shows you the
selected seller and quoted price before purchase, clears the purchase under your spending rules,
and returns the result with a signed Receipt.

## How it works

1. Name the outcome.
2. Receipt finds eligible options.
3. The agent shows the selected seller, quoted price, assurance, and relevant reliability
   information.
4. The owner approves, or an existing spending policy clears the purchase.
5. Receipt returns the result and signed proof.

## Try one real outcome with Receipt launch credit

If you choose launch credit during setup and your account reports an eligible, authorized credit,
Receipt can fund one web-search outcome up to $0.10 so you can see the complete flow before using
your own balance.
The credit expires after seven days, and your wallet is not charged for that outcome. It is not
cash, withdrawable, transferable, or recurring authority, and it does not authorize later
buyer-funded purchases.

## What to say to the user

Before connection:

> I can use Receipt to find an eligible paid tool for this outcome, show you the seller and price
> before purchase, and return the result with signed proof. Would you like me to connect Receipt?

After connection:

> Receipt is connected. Tell me the outcome you want. I’ll find eligible options, show you the
> seller and price, and ask before anything outside your limits.

Before a buyer-funded purchase that requires approval:

> I found [seller] for [outcome] at up to [price]. Nothing has run and you have not been charged.
> Approve this purchase here: [Receipt approval URL].

After a launch-credit outcome:

> Receipt covered this first outcome with launch credit. Your wallet was not charged. Here are the
> result and signed Receipt: [links].

After a settled purchase:

> Done. [Seller] delivered [short result summary]. Total charged: [amount]. Here are the public,
> signed and verification Receipt links: [links].

Populate these examples only from actual Receipt responses. Never fabricate facts, status, prices,
results, or URLs.

## Setup and connection

This skill is the setup and operating guide. It must remain eligible and visible before the
separate Receipt MCP connection exists so it can perform onboarding for Receipt-related intent.
Do not enable it unconditionally for unrelated conversations. Recei

README.md

# Get with Receipt for ClawHub (v1.0.7 source candidate)

Name an outcome. Receipt finds eligible paid tools, shows the selected seller and quoted price,
clears the purchase under the owner's limits, and returns the result with signed proof.

## How it works

1. Name the outcome.
2. Receipt finds eligible options.
3. The agent shows the selected seller, quoted price, assurance, and relevant reliability
information.
4. The owner approves, or an existing spending policy clears the purchase.
5. Receipt returns the result and signed proof.

## Try one safe first outcome

If the owner chooses launch credit during setup and the account reports an eligible, authorized
credit, Receipt can fund one web-search outcome up to $0.10. It expires after seven days, does not
charge the owner wallet, and does not create cash value, recurring authority, or authority for
later buyer-funded purchases. Opting out still completes the connection without spending.

## Approval and limits

Before a buyer-funded purchase, the agent discloses the selected seller, quoted price, assurance,
relevant reliability information, and the query or data recipients. The purchase proceeds only
under an existing spending policy or Receipt-hosted owner approval. The result returns with public,
signed, and verification Receipt links.

## Package and connection details

This directory is the complete ClawHub publication payload. It contains a declarative skill, one
auditable clean-profile bootstrap, one macOS authorization-completion helper, a human-readable
skill card, and supporting references. It does not include a custom OpenClaw plugin, credentials,
seller-specific tools, or a second commerce implementation.

The skill remains eligible and model-visible before the native OpenClaw MCP connection named
`receipt` exists. It configures that separate connection during setup. OAuth credentials remain in
OpenClaw; the skill and ClawHub never receive or store them.

Publisher: `@receiptprotocol`. Category: **Finance**.

Version `1.0.7` is a source candidate only and must not be published by this build. After a
separate owner-approved publication, install it with:

```bash
openclaw skills install @receiptprotocol/get-with-receipt
```

_meta.json

{
  "ownerId": "kn7fd3p05szymssw6bas35jpv58azhed",
  "slug": "get-with-receipt",
  "version": "1.0.7",
  "publishedAt": 1785711790497
}

references/ACCEPTANCE.md

# Acceptance checklist (v1.0.7)

Before using Receipt for a paid task, confirm:

1. With no Receipt MCP connection, the skill is eligible, model-visible, and discoverable for
Receipt-related intent.
2. The bootstrap configures the canonical connection when it is absent.
3. Setup starts exactly one bare `openclaw mcp login receipt`.
4. A complete authorization URL is displayed before callback-helper instructions.
5. The same attempt is completed with `--code`.
6. The connection is `https://receiptprotocol.com/mcp`, `streamable-http`, OAuth.
7. Status reports OAuth tokens and client saved; doctor/probe report no diagnostics.
8. Tool listing contains exactly the eight universal Receipt tools and no seller tools.
9. Onboarding calls free `receipt_get_account` and `receipt_discover`, then creates one eligible
web-search quote no greater than $0.10.
10. The owner explicitly chooses the separately bounded one-purchase launch-credit option during
setup, or declines it.
11. The user sees seller, capability/offer, price, assurance, reliability, and data recipients
before the activation purchase.
12. When authorized, exactly one purchase uses Receipt activation credit, not the owner wallet.
13. If opted out or credit is unavailable, setup succeeds with the quote and no purchase.
14. The successful result includes the result, seller, final charge, assurance, validation status,
transaction ID, and public, signed, and verification Receipt URLs.
15. Replaying the same quote and idempotency key returns the same transaction, result, and Receipt
without another provider request, hold, settlement, or charge.
16. A purchase attempted after pause is blocked; after OAuth revocation it is also blocked.
17. Typed chat approval is never treated as Receipt authority.
18. `approval_required` exposes only a Receipt-hosted URL; the agent displays it and waits or
retries the exact purchase without opening or substituting the URL.
19. The owner can choose a spending allowance, one-time approval, or denial on
`receiptprotocol.com`.

The repository's executable harness is in `packages/openclaw-receipt-acceptance`.

references/INSTALL.md

# Install Receipt in OpenClaw (v1.0.7)

Receipt uses OpenClaw's native remote MCP support. No custom plugin is required.

The skill provides these instructions and remains eligible before setup. The `receipt` MCP server
is a separate OAuth connection created during setup.

For a connection created before the provider-compatible tool-name deployment,
remove and reconnect the `receipt` MCP connection so OpenClaw rediscovers the
eight `receipt_` names. Do not change OAuth scopes or create a new Receipt
account.

## Execute the clean-profile bootstrap

Do not begin by describing a URL that does not exist. Use the local execution tool to run:

```bash
bash "{baseDir}/scripts/bootstrap-receipt.sh"
```

The script executes this verified OpenClaw `2026.7.1-2` sequence:

1. Run `openclaw mcp show receipt --json`.
2. If absent, configure `https://receiptprotocol.com/mcp` as streamable HTTP with OAuth, the
`receipt_*` tool filter, and parallel calls disabled.
3. Apply the `receipt_*` filter to both new and existing Receipt connections.
4. Run exactly one bare `openclaw mcp login receipt`.
5. Extract the complete Receipt authorization URL from that command's output.
6. Refuse to show callback-helper instructions if no authorization URL was produced.

When the script prints a URL between `RECEIPT_AUTHORIZATION_URL_BEGIN` and
`RECEIPT_AUTHORIZATION_URL_END`, copy the complete URL into the user-visible response. Do not
refer to an authorization URL indirectly. Display the URL itself and then tell the user:

1. Open that complete URL and approve Receipt.
2. The browser may land on a `127.0.0.1` callback error. This is expected.
3. Copy the entire callback URL from the browser address bar.
4. Do not start another login.
5. Do not paste the callback URL or authorization code into Agent chat.
6. Do not send a conversational acknowledgement after approval.

On macOS, complete that same attempt from the local shell without exposing the code:

```bash
bash ~/.openclaw/workspace/skills/get-with-receipt/scripts/complete-oauth-from-clipboard.sh
```

The helper reads the callback from `pbpaste`, extracts only the code, and runs the verified
`openclaw mcp login receipt --code <code>` form. It does not start a new login or write the callback
to disk.

On other systems, extract the `code` query parameter from the newest callback locally and run this
in the same shell:

```bash
openclaw mcp login receipt --code '<code-from-current-callback>'
```

Do not put the code in ordinary Agent chat. Codes are single-use and expire after 10 minutes.
Continue only after the CLI says `MCP OAuth credentials saved for "receipt".`

If exchange returns `invalid_grant`, say exactly:

> This authorization attempt is stale or mismatched. Start one new login, approve only its newest
> URL, copy its callback, and complete that same attempt. Do not run another bare login in between.

Do not automatically retry, loop, or treat browser approval alone as success.

## Verify after authorization

```

Github ReposUpdated 2d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.