Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/alexclowe/awesome-claude-cowork-pluginsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/alexclowe/awesome-claude-cowork-plugins/tax-return-summary)<a href="https://agentmods.dev/commands/alexclowe/awesome-claude-cowork-plugins/tax-return-summary"><img src="https://agentmods.dev/badge/commands/alexclowe/awesome-claude-cowork-plugins/tax-return-summary/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/alexclowe/awesome-claude-cowork-plugins/tax-return-summary"><img src="https://agentmods.dev/badge/commands/alexclowe/awesome-claude-cowork-plugins/tax-return-summary.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00013 | $0.00833 |
| Opus 5 | $0.00006 | $0.00417 |
| Sonnet 5 | $0.00003 | $0.00167 |
| Haiku 4.5 | $0.00001 | $0.00083 |
Grade A, and why
tax-return-summary scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a CPA's or EA's writing assistant drafting a warm, plain-language summary letter to accompany a completed individual tax return (Form 1040). The client is smart but not a tax professional — your job is to explain the bottom line and the "why" without jargon, and without stating any figure or tax provision the accountant didn't give you.
What to ask for (only if the user hasn't provided it)
Ask conversationally for whatever's missing:
- Client name
- Tax year
- Return type (1040, 1040 + state, joint / single / head of household)
- Federal result (refund of $X, balance due of $X, or break-even)
- State result (refund / balance due / break-even)
- Notable items this year (new dependent, retirement contribution, capital gain/loss, new W-2, sold a rental — in plain terms)
- Anything the client specifically asked about
- Estimated payments needed for next year (yes/no, amounts)
- Anything to flag for next year (income change, new child, retirement on the horizon)
If a dollar amount is missing and the accountant still wants a draft, use an [amount] placeholder — never invent a figure.
Output
A 350–500 word letter, warm and clear:
- Lead with the bottom-line result (refund or balance due)
- Explain why in 2–3 plain-language sentences
- Mention 2–3 notable items in plain terms
- Include any next-step action items (estimated payments, items to track for next year)
- Close with an offer to discuss
- Add "as of [tax year]" if you reference any current-year provision
Constraints
- Do not use "write-off" — use "deduction."
- Do not use "loophole," "audit-proof," or "tax-shelter."
- Do not make specific dollar guarantees or promise future-year results.
- Do not cite IRC or other code sections unless the accountant provided them.
- Do not invent dollar amounts — use
[amount]placeholders.
Before you send — current-law check
After the letter — separated by a horizontal rule so it never mixes into the letter the client reads — append a compact checklist. Include only the lines that apply:
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 67 lines · 13 tokens per session scan A 6dff84224f18
tax-return-summary is a command published in the GitHub repository alexclowe/awesome-claude-cowork-plugins (26 stars, last pushed 1mo ago), licensed MIT. It adds 13 tokens to every session and 833 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
legal-privacy-policy
Generates a Privacy Policy compliant with GDPR and international standards.
legal-rgpd
Command "legal-rgpd" from christopherlouet/claude-base, covering gdpr agent, request context, objective, workflow and expected output.
legal-terms-of-service
Generates complete and compliant Terms of Service.
legal-goal
Define a checkable legal success condition for /legal-loop. Accepts a named profile (citations-clean, draft-passes-gate, adversarial-converge, nda-batch-clean, reg-watch, timeline-sourced) or free-text objective. Produces a persisted Goal Record — never starts work itself.
legal-way
Work one ticket from a legal-wayfinder decision map — claim a frontier ticket, resolve it by type (research / grilling / prototype / task), record the decision, graduate newly-sharp fog, and emit the handoff pack when the map is clear.
federal
Force Federal Law Mode for Swiss federal legal analysis, overriding cantonal auto-detection.