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 SpartanLabsXyz/simmer-sdk --skill autoresearchgit clone --depth 1 https://github.com/SpartanLabsXyz/simmer-sdkWrote 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/spartanlabsxyz/simmer-sdk/autoresearch)<a href="https://agentmods.dev/skills/spartanlabsxyz/simmer-sdk/autoresearch"><img src="https://agentmods.dev/badge/skills/spartanlabsxyz/simmer-sdk/autoresearch/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/spartanlabsxyz/simmer-sdk/autoresearch"><img src="https://agentmods.dev/badge/skills/spartanlabsxyz/simmer-sdk/autoresearch.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00058 | $0.01337 |
| Opus 5 | $0.00029 | $0.00668 |
| Sonnet 5 | $0.00012 | $0.00267 |
| Haiku 4.5 | $0.00006 | $0.00134 |
Grade A, and why
simmer-autoresearch 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.
How it starts
The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Simmer Autoresearch
Autonomous experiment loop for trading skill optimization: try ideas, keep what works, discard what doesn't, never stop.
Based on pi-autoresearch (MIT).
Tools
init_experiment— configure session (name, skill_slug, metric, unit, direction). Call again to re-initialize with a new baseline.run_experiment— runs skill command, times it, captures output.log_experiment— records result. UsekeepONLY when the primary metric improved vs the baseline.discardif worse or unchanged. Zero trades ormetric=0is no-signal —discard, neverkeep.crashif the skill failed.checks_failedif post-run validation failed.keepauto-commits via git; the others auto-revert. Always include secondarymetricsdict. State the before→after comparison indescription(e.g.,"entry_threshold 0.05→0.03; $12→$18 pnl, keep"). Optionally includeasi(Actionable Side Information) for structured diagnostics.backtest_experiment— replay historical trades against new config without live execution. Fast config tuning (seconds vs hours). Requires trades withsignal_data(SDK 0.9.17+).
Setup
- Pick a skill to optimize and a primary metric (usually P&L).
git checkout -b autoresearch/<skill>-<date>- Read the skill source code thoroughly — understand what it does before mutating.
- Write
autoresearch.md— session spec with goal, metrics, how to run, constraints. - Write
autoresearch.sh— single command that runs the skill and outputs results. - Commit both files.
init_experiment→ run baseline withrun_experiment→log_experiment→ start looping.
autoresearch.md template
# Autoresearch: <goal>
## Objective
<What we're optimizing and the workload.>
## Metrics
- **Primary**: <name> (<unit>, lower/higher is better)
- **Secondary**: <name>, <name>, ...
## How to Run
`./autoresearch.sh` — runs the skill for one cycle.
## Constraints
- Only modify files in <skill directory>
- Do not change SDK core code
- Sim venue only (no real money)
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 · 92 lines · 58 tokens per session scan A dd7e67760b87
simmer-autoresearch is a skill published in the GitHub repository SpartanLabsXyz/simmer-sdk (48 stars, last pushed yesterday), licensed MIT. It adds 58 tokens to every session and 1,337 once invoked, about $0.0003 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
moonpay-scout
Prediction market arbitrage & alpha scout. Searches Polymarket and Kalshi for the same event, runs cross-platform arb math (including fees), and ranks opportunities by profitability. Use when asked to "find arb", "scout markets", "find edge", or scan a specific topic across prediction markets.
find-fade-setup
Find and evaluate a fade (mean-reversion) setup on prediction markets with oddsrail: candidates, overshoot signal, book, walked cost, resolution criteria, Kelly size, dry-run order.
check-cross-venue-edge
Decide whether a Polymarket vs Kalshi price difference is a real edge or a settlement mismatch: comparevenues, settlementaudit, quotecost on both legs, resolution criteria.
settle-resolved
Turn resolved and hedged Polymarket positions back into USDC, gasless, through the operator's relayer key, with a dry-run read-back first.
daily-review
Daily review of open prediction-market exposure with oddsrail: positions, resting orders, fills, markets closing soon, builder attribution.
polymarket
Query Polymarket: markets, prices, orderbooks, history.