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 varrdinc/varrd --skill varrd-discovergit clone --depth 1 https://github.com/varrdinc/varrdWrote 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/varrdinc/varrd/varrd-discover)<a href="https://agentmods.dev/skills/varrdinc/varrd/varrd-discover"><img src="https://agentmods.dev/badge/skills/varrdinc/varrd/varrd-discover.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00044 | $0.00727 |
| Opus 5 | $0.00022 | $0.00364 |
| Sonnet 5 | $0.00009 | $0.00145 |
| Haiku 4.5 | $0.00004 | $0.00073 |
Grade A, and why
varrd-discover 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 8d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
VARRD Discover — Autonomous Edge Discovery
Use this skill when a user wants VARRD to autonomously discover trading edges without driving the conversation step-by-step.
Command
varrd discover "<topic>"
How It Works
Give it a topic and VARRD handles everything: generates a creative hypothesis from its market structure knowledge base, loads data, charts the pattern, runs statistical tests, and returns the trade setup if an edge is found. One command, one result.
Each call tests ONE hypothesis through the full pipeline. Call again for another — VARRD uses tangential idea propagation, so related concepts branch out from your seed idea.
Examples
varrd discover "momentum on grains"
varrd discover "mean reversion on crypto" --mode explore
varrd discover "volatility clustering on ES" --markets ES NQ
varrd discover "seasonal patterns in energy" --test-type backtest
Options
| Flag | Values | Default | Description |
|---|---|---|---|
--mode |
focused, explore |
focused |
focused = stay close to topic, explore = creative freedom |
--markets |
e.g. ES NQ CL |
auto | Focus on specific markets |
--test-type |
event_study, backtest, both |
event_study |
Type of statistical test |
When to Use discover vs research
| Scenario | Use |
|---|---|
| You have a specific idea to test | varrd research "your idea" |
| You want broad exploration of a space | varrd discover "topic" |
| You want control over each step | varrd research |
| You want hands-free, one command | varrd discover |
| Running many hypotheses at scale | varrd discover in a loop |
Reading the Output
The result includes:
- Edge verdict: STRONG EDGE / MARGINAL / NO EDGE
- Statistics: Win rate, Sharpe ratio, signal count
- Trade setup (if edge found): exact entry, stop-loss, take-profit prices
- Hypothesis ID: saved to your library for later scanning
Tips
- Use
--mode explorefor more creative, tangential ideas - Each call takes 30-120 seconds — it's running the full pipeline
- Run in a loop to test many hypotheses:
for topic in "momentum" "mean reversion" "breakouts"; do varrd discover "$topic"; done - All validated hypotheses are saved — use
varrd scanlater to see what's firing
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.
- 8d ago First seen · 80 lines · 44 tokens per session scan A 74fe6e9743f7
varrd-discover is a skill published in the GitHub repository varrdinc/varrd (24 stars, last pushed 7d ago), licensed MIT. It adds 44 tokens to every session and 727 once invoked, about $0.0002 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
jcdb
A read-only database lookup skill that checks real tables, columns, definitions, and selected data before SQL or code is written. Its documentation is in Chinese.
devwiki
Use when the user asks about JoyinCode development workflow, how to use AI coding in this project, or mentions any pipeline step (markitdown Word-to-MD conversion, grill-with-docs requirement analysis, opsx proposal workflow).
specx
An enterprise software-development workflow that takes a request through clarification, specification, design, task planning, execution, and archiving.
grill-with-docs
Full design grilling session. I interview you relentlessly about every aspect of your plan until we reach shared understanding. At the end, I generate CONTEXT.md and ADR files from what we discussed.
opsx
A command for managing proposals in the OPSX workflow. It supports starting, applying, archiving, and initialising proposals.
demo
A project-analysis skill for the cloudQuant/backtrader repository, with steps for reviewing its structure, related processes, and visible results.