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 carta/plugins --skill carta-interaction-reference-coregit clone --depth 1 https://github.com/carta/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/skills/carta/plugins/carta-interaction-reference-core)<a href="https://agentmods.dev/skills/carta/plugins/carta-interaction-reference-core"><img src="https://agentmods.dev/badge/skills/carta/plugins/carta-interaction-reference-core/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/carta/plugins/carta-interaction-reference-core"><img src="https://agentmods.dev/badge/skills/carta/plugins/carta-interaction-reference-core.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 | $0.00088 | $0.02258 |
| Opus 5 | $0.00044 | $0.01129 |
| Sonnet 5 | $0.00018 | $0.00452 |
| Haiku 4.5 | $0.00009 | $0.00226 |
Grade A, and why
carta-interaction-reference-core 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 — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Carta AI Agent Interaction Reference — Core
The rules an agent cannot get wrong: what it may state as fact, what it must label, what it must ask before doing, and what it must never do.
Section numbers match carta-interaction-reference, so a citation like §4.1 or §6.2 resolves the same in either file. That file is the superset — it adds voice and tone, vocabulary and audience calibration, proactiveness, error-handling style, and waiting-state humour.
2.3 Financial Precision {#ref-ext:financial-precision}
When presenting financial data, be exact. Carta is infrastructure for financial records — approximation erodes trust.
- Always include units: "$12,400" not "12,400." "1,200 shares" not "1,200."
- Always include dates: "as of March 31, 2026" — not "recently" or "the latest."
- Never round without disclosure: If displaying a rounded number, say so. "Approximately $4.2M (exact: $4,187,340.22)."
- Never fabricate data. If the agent doesn't have a number, say it's unavailable. Never fill in a plausible value (see [Section 6.1][ref-ext:no-fabrication]).
- Label estimates as estimates. Some financial data is inherently imprecise — 409A valuations, pro forma models, projected distributions. Always state the methodology, source, and as-of date. Example: "The estimated fair market value is $4.12/share (409A valuation, backsolve method, as of March 1, 2026)." Never present an estimate with the same confidence as an exact record.
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 · 141 lines · 88 tokens per session scan A 10fd141a4d8d
carta-interaction-reference-core is a skill published in the GitHub repository carta/plugins (13 stars, last pushed today), licensed Apache-2.0. It adds 88 tokens to every session and 2,258 once invoked, about $0.0004 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 2026-09-02.
Other skills, from other repositories
bionic-safety-net
Unified survival infrastructure: health, finance, legal safety, circuit breakers, and structural protection against all ruin classes. The last line of defense.
insurance-expert
Expert-level insurance systems, underwriting, claims processing, actuarial analysis, risk assessment, and insurtech solutions. Use when the user mentions underwriting, claims, actuarial, risk, or insurtech, or when the task involves Insurance Systems, Insurance Types, Standards and Regulations, or Claims Processing.
banking-expert
Expert-level banking systems, core banking, regulations, and banking technology. Use when the user mentions finance, core banking, regulations, or payments, or when the task involves Banking Systems or Key Technologies.
regtech-expert
Expert in regulatory technology, compliance automation, KYC/AML, transaction monitoring, risk assessment, and automated reporting. Use when the user mentions compliance, KYC, AML, transaction monitoring, risk assessment, or regulatory reporting, or when the task involves KYC/AML Compliance, KYC/AML Program, Regulatory…
gcp-dispute-kit
Use to assemble a dispute-grade evidence packet and ready-to-file letters after GCP API-key fraud (leaked-key abuse, runaway Gemini/Vertex charges). READ-ONLY — mutates nothing. Triggers on "dispute gcp charges", "google cloud fraud refund", "gemini api fraud dispute", "gcp billing dispute", "chargeback google cloud".
thai-invoice
Use this skill for any task involving Thai tax invoices, receipts, quotations, credit/debit notes, or withholding-tax (WHT) certificates. Trigger on "ออกใบกำกับภาษี", "ใบเสร็จ", "ใบเสนอราคา", "ใบลดหนี้", "ใบเพิ่มหนี้", "หนังสือรับรองหักภาษี ณ ที่จ่าย", "ภ.ง.ด.3", "ภ.ง.ด.53", "Thai tax invoice", "VAT 7%", "withholding…