Tip with Grove

WarnAudited by ClawScan on May 10, 2026.

Overview

This is a coherent Grove tipping guide, but it enables real-money tipping and funding with autonomous/cron modes and local wallet or API-key authority that users should review carefully.

Install only if you intentionally want an agent-capable real-money tipping tool. Use a dedicated low-balance wallet or limited API key, review the remote installer, avoid unattended cron/--yes automation unless you set strict budgets, and keep webhook/log destinations private.

Findings (6)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If an agent or automation runs this with --yes, it can spend from the configured wallet without a fresh human confirmation for that funding action.

Why it was flagged

The script can move funds into the Grove account while skipping confirmation. Because this is a real-money action, the confirmation-bypass path needs explicit user budgeting and operational controls.

Skill content
--yes                      Skip confirmation prompt ... fund_result=$(grove fund "$FUND_AMOUNT" --network "$NETWORK" --json 2>&1)
Recommendation

Use dry-run and manual confirmation by default; only enable --yes with strict wallet funding limits, monitoring, and an explicit spending budget.

What this means

A misconfigured or forgotten cron job could keep topping up the Grove account whenever the balance falls, potentially moving more funds than expected over time.

Why it was flagged

The documentation explicitly shows a persistent cron job that runs every 15 minutes with confirmation disabled, creating unattended recurring financial automation.

Skill content
Typical cron usage:
    */15 * * * * /path/to/auto-fund.sh --yes >> /var/log/grove-auto-fund.log 2>&1
Recommendation

Do not enable the cron example unless you have a capped wallet, clear spending limits, alerting, and a plan to disable it.

What this means

Installing and using the skill may involve giving the agent access to a funded wallet or Grove API key capable of sending tips or funding the account.

Why it was flagged

The skill expects payment credentials and wallet authority for autonomous agents, but the registry metadata declares no primary credential, making the privilege requirements easy to underappreciate.

Skill content
Wallet + Crypto (recommended) | Autonomous agents | Creates wallet, fund with USDC + ETH on Base, self-funds via x402 ... I have an API key ... paste API key
Recommendation

Use a dedicated low-balance wallet or limited API key, avoid sharing personal wallets, and confirm what permissions the Grove CLI stores under ~/.grove.

What this means

You are trusting the current contents of Grove’s remote installer at install time.

Why it was flagged

The CLI installation is a remote shell script piped to bash. This is disclosed and purpose-aligned, but the provided artifacts do not include a checksum, pinned version, or reviewed installer contents.

Skill content
curl -fsSL https://grove.city/install-cli.sh | bash
Recommendation

Review the installer before running it, install from a pinned release if available, and only use it if you trust grove.city.

What this means

A webhook recipient could see Grove balance alerts and related timing information.

Why it was flagged

The monitor can send balance and threshold details to a user-supplied webhook. This is expected for alerts, but it transmits financial balance information outside the local machine.

Skill content
curl -X POST -H "Content-Type: application/json" \
             -d "$payload" "$WEBHOOK_URL"
Recommendation

Use only trusted webhook URLs, avoid sending alerts to shared or public endpoints, and rotate webhook secrets if exposed.

What this means

Your agent logs may contain a history of what content you tipped, how much you paid, and why.

Why it was flagged

The skill recommends persisting tip metadata in agent logs. That is purpose-aligned for Grove’s knowledge-graph concept, but it records payment-related history and rationale for later reuse.

Skill content
Recommended fields (store in your agent logs): Canonical URL ... Destination ... Amount + token/network ... rationale
Recommendation

Keep those logs private, set retention limits, and avoid logging sensitive browsing or payment rationale unless needed.