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 sophymarine/openregistry --skill find-companygit clone --depth 1 https://github.com/sophymarine/openregistryWrote 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/sophymarine/openregistry/find-company)<a href="https://agentmods.dev/skills/sophymarine/openregistry/find-company"><img src="https://agentmods.dev/badge/skills/sophymarine/openregistry/find-company/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/sophymarine/openregistry/find-company"><img src="https://agentmods.dev/badge/skills/sophymarine/openregistry/find-company.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 | $0.00097 | $0.00546 |
| Opus 5 | $0.00048 | $0.00273 |
| Sonnet 5 | $0.00019 | $0.00109 |
| Haiku 4.5 | $0.00010 | $0.00055 |
Grade A, and why
find-company 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.
What it actually says
Find Company
Confirm a company exists and capture its core identity. Stop there.
When to use
- "Does X exist as a real company?"
- "Is X still active?"
- "What's the registry number for Acme Trading Ltd?"
- "Verify this counterparty before I sign."
When NOT to use
- Need directors →
get-director-details - Need ownership →
get-shareholder-details - Need financial figures →
get-financials - Need filing history →
get-filings
Example
Is "Tesla Limited" a real UK company?
Verify SIREN 552081317.
Workflow
1. If user gave a name:
search_companies({ jurisdiction, query, limit: 5 })
→ pick the candidate whose status='active' and name best matches
→ if multiple plausible matches, surface them and ask the user to confirm
2. If user gave an ID, skip step 1.
3. get_company_profile({ jurisdiction, company_id })
→ capture company_name, status, status_detail,
incorporation_date, registered_address
4. REPORT:
- 1 line: "<name> (<jurisdiction>, <company_id>) is <status>, incorporated <date>."
- Address on second line if user asked.
- Cite the company_id verbatim.
Edge cases
- Multiple matches: don't guess. Ask the user to confirm.
- Dissolved: report as such; suggest
get-filingsorget-director-detailsif the user wants the historical view. - 501 from upstream: this jurisdiction isn't supported. Surface honestly.
- Per-country ID format: differs per registry. Call
list_jurisdictions({ jurisdiction: '<CC>' })for the schema.
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 · 63 lines · 97 tokens per session scan A f5b6185f0769
find-company is a skill published in the GitHub repository sophymarine/openregistry (17 stars, last pushed 4d ago), licensed Apache-2.0. It adds 97 tokens to every session and 546 once invoked, about $0.0005 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 skills, from other repositories
corporate-ownership-investigation
Trace who ultimately owns or controls a company, screen it for sanctions, and report the result without overclaiming. Use when asked "who owns X", "who is behind X", "is X sanctioned", "is X connected to Y", "what does X own", or when doing KYC/AML/due-diligence, counterparty checks, supply-chain screening, or…
qcc-executive-background
A Chinese company and executive background-check workflow examines a named person together with a company's Unified Social Credit Code, the identifier used for Chinese businesses.
qcc-risk-screen
A full risk-screening workflow for Chinese companies using Qichacha, a business-information service. It checks legal, enforcement, tax, bankruptcy, ownership, operating, and other company-risk records.
qcc-basic-profile
A workflow for Qichacha, a Chinese company-information service, that gathers a basic business summary in one pass. It covers registration, contacts, key people, tax and invoice details, branches, listings, and annual reports.
qcc-ipr-portfolio
A Chinese-language workflow for listing a company's intellectual-property and online assets. It covers patents, trademarks, copyrights, integrated-circuit layouts, standards, licenses, websites, apps, online shops, and social-media accounts.
qcc-ownership-trace
A workflow for tracing a company's ownership, including its controlling person, direct shareholders, investments, beneficial owners, and registration changes.