← Back to Skills Marketplace
Taxguard Skill
by
aaronahadi23
· GitHub ↗
· v1.1.0
· MIT-0
116
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install taxguard-skill
Description
Monitors every trade for tax risks and optimization silently, logs results, blocks trades if Guardian Mode enabled, and delivers daily tax reports.
Usage Guidance
This skill largely does what it says (remote tax checks and optional blocking), but before installing you should: 1) Verify the provider (rhetra/api.rhetra.io) and their privacy/terms—there's no homepage in the metadata. 2) Recognize the skill will transmit detailed, sensitive financial and tax data (positions, recent sales, MAGI, filing status) to a third-party API on every trade; only proceed if you trust that endpoint and its security/privacy practices. 3) Note the manifest says no credentials required but the scripts require an API key—treat this mismatch as a red flag and confirm where/how keys are stored. 4) Consider whether 'silent' monitoring (not showing results to the trader) is acceptable for your use case; require explicit user consent if deployed for real accounts. 5) Test on a paper/pseudo account first and consider hosting an internal review or code audit of the scripts (they are readable but perform network calls). If you need higher assurance, request a public homepage, documentation, company identity, privacy policy, and an option to self-host or run checks locally without sending MAGI/other PII externally.
Capability Analysis
Type: OpenClaw Skill
Name: taxguard-skill
Version: 1.1.0
The skill functions as a 'man-in-the-middle' for all trading activities, instructing the agent in SKILL.md to silently transmit sensitive financial data—including account equity, full position history, and estimated MAGI—to an external API (api.rhetra.io) before every trade. While the data requested is relevant to tax analysis, the explicit instructions to perform these checks 'SILENTLY' and 'DO NOT interrupt the trader' effectively hide the transmission of sensitive PII and financial context from the user. This behavior, implemented in check-trade.js and signup.js, represents a high-risk data exfiltration surface despite the lack of overt credential theft or malicious code execution.
Capability Assessment
Purpose & Capability
Name/description (silent trade checks, daily report, optional blocking) aligns with the shipped scripts: signup.js obtains an API key and check-trade.js posts trade context to a remote API for decisions. No unrelated cloud credentials or binaries are requested.
Instruction Scope
SKILL.md instructs the agent to gather very sensitive financial and tax context (positions, recent sales, equity, day-trades, MAGI, filing status, cost-basis, etc.) and to 'do NOT show the results to the trader' (silent monitoring). All of that data is sent to a remote endpoint for every check. Hiding results from the user is an ethical/privacy concern and expands the skill's authority in a way users should explicitly consent to.
Install Mechanism
No install spec or remote downloads; the skill is instruction-plus-local scripts only. That minimizes supply-chain risk. The shipped JS files make outbound HTTPS requests to api.rhetra.io which is expected for a remote service.
Credentials
The registry metadata declares no required env vars/credentials, but SKILL.md and signup/check-trade scripts expect and use an API key (TAXGUARD_API_KEY or --key). The skill asks for extensive sensitive tax/account data (MAGI, positions, recent sales). Those fields are plausible for tax computation, but they are sensitive and the mismatch in declared vs. actual credential requirements is an inconsistency worth flagging.
Persistence & Privilege
The skill is not 'always' enabled and does not request system-wide config changes or modifications to other skills. It behaves as a normal, on-demand service that can be invoked by the agent.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install taxguard-skill - After installation, invoke the skill by name or use
/taxguard-skill - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
**New monitoring and reporting workflow for every trade, with daily tax summaries.**
- All trades are pre-checked silently by TaxGuard for taxable events before execution; results are logged, not shown, unless Guardian Mode triggers a block.
- Daily tax report summarizes flagged events (wash sales, tax-saving opportunities, estimated tax), delivered automatically at end of day or on trader request.
- Guardian Mode can block trades with specific risks (e.g., wash sales, PDT), requiring explicit confirmation from the trader to proceed.
- Setup is streamlined with instant free API key registration and test check; supports multiple brokerages and crypto exchanges.
- Trader can access real-time summaries, export reports for accounting, or enable/disable blocking at any time with simple commands.
Metadata
Frequently Asked Questions
What is Taxguard Skill?
Monitors every trade for tax risks and optimization silently, logs results, blocks trades if Guardian Mode enabled, and delivers daily tax reports. It is an AI Agent Skill for Claude Code / OpenClaw, with 116 downloads so far.
How do I install Taxguard Skill?
Run "/install taxguard-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Taxguard Skill free?
Yes, Taxguard Skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Taxguard Skill support?
Taxguard Skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Taxguard Skill?
It is built and maintained by aaronahadi23 (@aaronahadi23); the current version is v1.1.0.
More Skills