veris
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BRAVE_API_KEY | No | Your Brave Search API key for better search results. Leave empty for keyless DuckDuckGo fallback. | |
| SEC_USER_AGENT | No | Your name and email for SEC fair-access policy (e.g., 'Your Name you@email.com'). Defaults to a generic agent. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| web_searchA | Search the web. Returns ranked results (title, url, snippet) as structured JSON. Uses Brave Search when BRAVE_API_KEY is set, otherwise keyless DuckDuckGo. |
| web_readA | Fetch a URL and return clean markdown PLUS provenance: published date, author, canonical URL, content hash, fetch time, and license signals. The provenance is the point — verifiable source metadata an AI normally cannot get. |
| web_researchA | One-shot research: search the web, fetch the top results, and return their clean content bundled with full provenance per source. Ideal for grounding an answer with citations. |
| finance_filingsA | List a company's recent SEC filings from EDGAR by ticker, name, or CIK. Free, official, no API key. Returns form type, filing/report dates, accession number, and direct document URL — provenance is authoritative (straight from SEC). |
| finance_filing_readA | Fetch and clean a specific SEC filing. Pass a filing document URL, OR a company query plus formType to auto-read the most recent matching filing (e.g. latest 10-K for NVDA). Returns clean text plus provenance. |
| finance_financialsA | Key structured financials (revenue, net income, total assets, cash, diluted EPS) from SEC XBRL data. Each figure is stamped with the exact filing it came from (form, filed date, accession, fiscal period) — authoritative provenance. Ticker / name / CIK. No API key. |
| watch_manageA | Add, remove, or list watches. Watch a company's SEC filings (target = ticker/name/CIK, optional formType like 8-K) or any URL (target = https://…, watched by content hash). Adding baselines current state; use watch_check later to see only what's new. By default only MATERIAL forms alert (8-K, 10-K/Q, 13D/G, mergers, offerings, late filings) — routine Form 4 / 13F noise is suppressed. Pass formType="ALL" to alert on every form. |
| watch_checkA | Check every watch for changes since the last check: new SEC filings, changed page content. Returns only what's new and rolls baselines forward. Routine forms (Form 4, 13F, etc.) are suppressed unless the watch asked for them. Run it on a schedule (cron / scheduled agent) to turn veris into an alert feed. Note: the SEC filing index may lag up to ~1h (cache). |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
Most tools target a clearly distinct action: watch_manage (mutate watches) vs watch_check (poll changes), and finance_filings (list) vs finance_filing_read (read one filing) are well separated. The main overlap is web_research, which is essentially web_search + web_read bundled, so an agent may wonder when to use it versus the two primitives.
A clean domain-prefix scheme (watch_, finance_, web_) is applied throughout, which aids grouping. However the action portion is inconsistent: watch_manage and web_read are prefix+verb, while finance_filings and finance_financials are prefix+noun and finance_filing_read is prefix+noun+verb, so the finance cluster breaks the pattern.
Eight tools is a tight, well-scoped set for a filings-monitoring plus web-research server. Each tool earns its place and none feels redundant or padded.
The watch lifecycle (add/remove/list + check) and the SEC surface (list filings, read filing, structured financials) plus web search/read/research give solid coverage. Minor gap: there's no explicit way to inspect or clear accumulated alert history beyond watch_check's rolling output.