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 krakenfx/kraken-cli --skill kraken-lab-experimentgit clone --depth 1 https://github.com/krakenfx/kraken-cliWrote 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/krakenfx/kraken-cli/kraken-lab-experiment)<a href="https://agentmods.dev/skills/krakenfx/kraken-cli/kraken-lab-experiment"><img src="https://agentmods.dev/badge/skills/krakenfx/kraken-cli/kraken-lab-experiment/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/krakenfx/kraken-cli/kraken-lab-experiment"><img src="https://agentmods.dev/badge/skills/krakenfx/kraken-cli/kraken-lab-experiment.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 | $0.00061 | $0.02731 |
| Opus 5 | $0.00030 | $0.01366 |
| Sonnet 5 | $0.00012 | $0.00546 |
| Haiku 4.5 | $0.00006 | $0.00273 |
Grade A, and why
kraken-lab-experiment 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 9d 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 — 252 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lab Experiment Lifecycle
PREREQUISITE: Load
kraken-playgroundto understand what a session is and where it lives.
Turn a trading hunch into a mechanical verdict: pre-register the hypothesis and its success criteria (hash-sealed, so the goalposts cannot move), run it as recorded paper runs, score each session against its own tape, and let the frozen criteria — not vibes — say pass or fail. Every step reads and writes plain files under the active scope, so the lifecycle survives kills, restarts, and context loss: whatever step you find half-done on disk is the step you resume.
Use this skill for:
- driving one experiment end-to-end under
/loop(freeze → run → score → verdict) - resuming an experiment a previous context started
- reproducing a verdict cold, from files alone
Safety rules
- Paper only: run experiments inside a paper workspace
(
export KRAKEN_WORKSPACE=<name>), where the order verbs fill on the paper account and no real funds are reachable. - Never edit a frozen experiment file. The seal exists to catch exactly
that; a tampered spec fails every later step with a
parseenvelope. - Never report a verdict you did not obtain from
lab score --experimentorlab compareoutput in this session. No estimating, no "it would probably pass".
The state machine
With a sealed session plan the whole table collapses to one command — run
kraken lab next <exp> and do what it says:
kraken lab next momentum-1 -o json 2>/dev/null
# state: "start" → run the exact `command` it prints, then stop+score
# state: "running" → drive or wait, then `kraken session stop` + `lab score`
# state: "plan_complete" → `kraken lab compare <exp>`, conclude (Step 5)
lab next re-derives progress from the sessions tree on every call: completed
runs consume plan entries in order, aborted sessions keep their ordinals and
count for nothing, and a replay entry whose tape no longer matches its
sealed hash is refused outright. Stop and score a finished run before
asking again — an unstopped run has no summary and reads as aborted.
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.
- 9d ago First seen · 252 lines · 61 tokens per session scan A 373500d91962
kraken-lab-experiment is a skill published in the GitHub repository krakenfx/kraken-cli (711 stars, last pushed 1mo ago), licensed MIT. It adds 61 tokens to every session and 2,731 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
sector-rotation
An analysis framework for comparing industries in the Chinese A-share stock market, using business conditions, price momentum, valuation, and money flows. It produces rankings and higher- or lower-allocation suggestions.
strategy-pivot-designer
Detect backtest iteration stagnation and generate structurally different strategy pivot proposals when parameter tuning reaches a local optimum.
twitter-reader
Read Twitter/X for financial research using opencli (read-only). Use this skill whenever the user wants to read their Twitter feed, search for financial tweets, view bookmarks, look up user profiles, or gather market sentiment from Twitter/X. Triggers include: "check my feed", "search Twitter for", "show my…
chenhao-limit-up
A framework for judging Chinese A-share stocks that have reached the daily price-rise limit, using market mood, sector leadership, and trading momentum.
trading-risk-gate
Unified pre-trade safety gate: Ruin check (Law #1), ergodicity audit, and win-rate dominance validation. Absorbs: ergodicity-check, law-of-ruin, win-rate-dominance.
vectorbt
High-performance vectorized backtesting with parameter optimization, portfolio simulation, and rich performance metrics.