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.
npx skills add deciqAI/knowledge-skills --skill cpa-irs-representationgit clone --depth 1 https://github.com/deciqAI/knowledge-skillsWrote 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/skills/deciqai/knowledge-skills/cpa-irs-representation)<a href="https://agentmods.dev/skills/deciqai/knowledge-skills/cpa-irs-representation"><img src="https://agentmods.dev/badge/skills/deciqai/knowledge-skills/cpa-irs-representation/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/skills/deciqai/knowledge-skills/cpa-irs-representation"><img src="https://agentmods.dev/badge/skills/deciqai/knowledge-skills/cpa-irs-representation.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 | $0.00172 | $0.03427 |
| Opus 5 | $0.00086 | $0.01714 |
| Sonnet 5 | $0.00034 | $0.00685 |
| Haiku 4.5 | $0.00017 | $0.00343 |
Grade A, and why
cpa-irs-representation 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 10d 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CPA/EA — IRS Representation
Not legal or tax advice. No deadlines, thresholds, dollar limits, eligibility criteria, penalty amounts or interest rates appear here — all of them expire or are matter-specific. The notice states its own date. Verify limitation periods and eligibility against current authority every time. Where facts suggest exposure beyond a civil dispute, involve counsel.
Activate when: a notice arrives; an exam opens; unfiled returns or an unpaid balance exists; enforcement is pending; an appeal decision is live; you are scoping or handing off a representation engagement. Do NOT activate when: there is no controversy, or the matter belongs to counsel.
Why this skill
Representation punishes improvisation. Cases go badly not on the merits but by responding to the wrong notice, missing a date printed on page one, conceding scope nobody asked to expand, or putting a client into a resolution they were never going to sustain.
Structure: an intake gate, two fact phases, four routes. The gate exists because the most consequential decision in representation is whether to take the case at all.
Process
Gate — Before you take the case
Each item can end the engagement, and each is cheaper to check now than later.
- Authorized? Representation rights differ by credential and by whether you prepared the return. Confirm your standing for this taxpayer, these tax types, these periods.
- Conflict? Common and easy to miss: spouses whose interests diverge once liability allocation is live; an entity and its owner where the adjustment shifts between them; two shareholders on one K-1 issue; a client whose position implicates a return you prepared. Where consent is permissible, get it in writing before the first adverse development.
- Competent for this matter? Competence is matter-specific. Having prepared the return does not make you the right person to defend it. Referring out is a better professional outcome than learning on a client's case.
- Did you prepare the return under exam? Not disqualifying, but your work is now part of the record and your judgment about your own position is not neutral. Name it to yourself, and consider whether the client should hear it.
- Scope, authorization and fee in writing. Which tax types, which periods, which matters — the authorization form is a scope document, and a broad one grants more than the client may intend. Preparing unfiled returns is usually a separate engagement from representing on the balance. Define how authority gets withdrawn at the end.
- Privilege relevant? Protection for communications with a federally authorized practitioner is limited and does not extend everywhere — notably not to criminal matters and not to return preparation.
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.
- 10d ago First seen · 175 lines · 172 tokens per session scan A cefd8b1d81be
cpa-irs-representation is a skill published in the GitHub repository deciqAI/knowledge-skills (10 stars, last pushed 8d ago), licensed MIT. It adds 172 tokens to every session and 3,427 once invoked, about $0.0009 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 2026-08-31.
Other skills, from other repositories
review-contract
Clause-by-clause contract review — inventories key clauses (term, termination, liability cap, indemnity, IP, payment, confidentiality, governing law, auto-renewal), flags deviations from market-standard positions with RED/YELLOW/GREEN severity and verbatim quotes, drafts redline language, and sets negotiation…
fiscaliste
Fiscaliste IA pour la fiscalité personnelle des particuliers français : optimisation et déclaration de l'impôt sur le revenu, IFI, revenus du capital, revenus fonciers, equity salarial, crypto-actifs et PER. Couvre le calcul de l'IR (barème, quotient familial, décote, PAS, CEHR, revenus exceptionnels), la déclaration…
commissaire-aux-comptes
Commissaire aux comptes IA pour l'audit des comptes annuels d'entreprises françaises. Applique la démarche NEP en 7 phases : prise de connaissance, contrôle du FEC, vérification du bilan, du compte de résultat, de la balance, de la liasse fiscale, et contrôles transversaux. Émet une opinion motivée sur la fiabilité…
agb-bei-kreditvertraegen-verbraucherdarlehen
Für AGB bei Kreditverträgen Verbraucherdarlehen: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt.
tax
Structures the tax questions a growing business faces — corporate income, sales and use, payroll, nexus, and the obligations created by hiring or selling somewhere new. Use this to work out what a new state or country obligates you to, prepare for a tax filing or audit, understand sales tax on your product, or check…
802-regulations-dora
Use when reviewing, designing, or modifying Java enterprise systems that may support financial entities, critical ICT services, third-party ICT provider integrations, or operational resilience obligations under DORA. This should trigger for requests such as Review a Java platform for DORA ICT risk controls; Design…