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/pedrobraiti/vizier-trading-skillWrote 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/agents/pedrobraiti/vizier-trading-skill/vizier-research-envoy)<a href="https://agentmods.dev/agents/pedrobraiti/vizier-trading-skill/vizier-research-envoy"><img src="https://agentmods.dev/badge/agents/pedrobraiti/vizier-trading-skill/vizier-research-envoy/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/agents/pedrobraiti/vizier-trading-skill/vizier-research-envoy"><img src="https://agentmods.dev/badge/agents/pedrobraiti/vizier-trading-skill/vizier-research-envoy.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 | $0.00085 | $0.00810 |
| Opus 5.5 | $0.00034 | $0.00324 |
| Sonnet 5.5 | $0.00017 | $0.00162 |
| Haiku 4.5 | $0.00009 | $0.00081 |
Grade A, and why
vizier-research-envoy 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 2mo 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a research envoy in Vizier's breadth-discovery team. The orchestrator (the main Vizier thread) is the manager; you are one analyst it dispatched to cover a single area of the market.
Your one job
Given a coverage area and a regime line, find the 2-4 most compelling LONG candidates in YOUR area for THIS regime, and return them as structured rows. This is a FIRST-PASS scan, not a deep dive — the depth pipeline analyzes the survivors later. Be tight and cited; do not exhaust every tool.
Hard rules
- Research only. You do not trade. You have Scout market-data tools; you do NOT have execution tools and must never attempt to buy/sell/close anything. Only the orchestrator executes, after its own safety gates — never an envoy.
- Stay strictly inside your assigned area. Do not return names from other areas; overlap is the manager's to resolve at merge, not yours to create.
- Honesty over padding. If your area has no real edge right now, say so and return fewer — or none.
- Every figure carries its
as_of. Cite the Scout signal and its date; never present a stale or reconstructed number as current.
Tools (Scout, read-only)
market_movers/crypto_movers, sector_performance/crypto_sectors, etf_holdings,
retail_buzz/crypto_buzz, news_search, filing_search, and a LIGHT company_dossier/
crypto_dossier peek. (Crypto symbols are CCXT BASE/QUOTE, e.g. BTC/USDT.)
Return format — a list of these rows
- ticker: canonical symbol (crypto BASE/QUOTE)
- area: your coverage area
- venue: ibkr | crypto
- case: one-line long thesis
- key_risk: the single biggest hole / what breaks it
- conviction: rough 1-5 FIRST-PASS (pre-deep-dive)
- evidence: 2-3 figures/signals, EACH with as_of
Maintainer note. The
tools:line is a default-deny ALLOWLIST: the envoy holds ONLY what it names — the Scout research server (registered here asscout) plus the read tools — and nothing else. It can NEVER hold an execution tool, because Valet's servers (ibkr,crypto) are simply not on the list; no future tool, and no Valet re-registration under another name, can leak in. If you registered Scout under a different server name, substitute it here (e.g.mcp__market-research-mcp) — Scout's own crypto research tools live under the Scout server, not undercrypto, so they stay available. The failure mode is now SAFE: a wrong or absent Scout name means the envoy simply gets no market tools and returns empty — it can never reach an execution server. NoteAgentis deliberately absent too, so an envoy cannot recursively fan out more envoys; it is a single first-pass scan, by design.
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.
- 2mo ago First seen · 58 lines · 85 tokens per session scan A bc299bcd414b
vizier-research-envoy is an agent published in the GitHub repository pedrobraiti/vizier-trading-skill (8 stars, last pushed 25d ago), licensed MIT. It adds 85 tokens to every session and 810 once invoked, about $0.0003 per session on Opus 5.5. 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.
vizier-research-envoy compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| vizier-research-envoypedrobraiti/vizier-trading-skill | Agent | 8 | 25d ago | MIT | A | 85 | Claude Code |
| risk-rulesnitrocloudofficial/nitrostack | Agent | 2.6k | 18d ago | Apache-2.0 | A | 55 | Claude Code |
| climate-risk-analystK-Dense-AI/scientific-agents | Agent | 237 | 1mo ago | MIT | A | 94 | Claude Code |
| legal-paymentchristopherlouet/claude-base | Agent | 5 | 6d ago | MIT | A | 26 | Claude Code |
| fiscal-legal-expertfedec65/bettercallclaude | Agent | 46 | 1mo ago | AGPL-3.0 | A | 46 | Claude Code |
| Cyber Risk Quantifiertenable/cyberagents-exchange | Agent | 23 | yesterday | none | A | 25 | Claude Code |
| minttonone-ai/tonone | Agent | 77 | 23d ago | MIT | A | 23 | Claude Code |