desk-strategy-lab SKILL.md skillA

desk-strategy-lab from swcstudiospace/pumpgrok is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add swcstudiospace/pumpgrok/desk-strategy-lab@git:20261023.ef2f2f6

Pins git:20261023.ef2f2f6 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
swcstudiospace/pumpgrok
Path
skills/desk-strategy-lab/SKILL.md
Versions recorded
2
Latest version
git:20261023.ef2f2f6, committed Oct 23, 2026
First version recorded
Aug 26, 2026
Size
1,621 bytes
Repository stars
42
License
AGPL-3.0

Lightweight rules and paper-trading support for entry filters, position sizing, and exit logic. Allows the desk to experiment without touching live capital until rules are proven.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Desk Strategy Lab

## L1 - Summary
The sandbox for entry filters, position sizing, and exit logic. Experiments run against paper fills only, so ideas succeed or fail on evidence without risking capital.

## L2 - Method
1. Paper mode is the default; rule changes go through `replay-and-backtest` first:
```bash
pumpgrok run --mode paper --replay tests/fixtures/replay/<dataset>
```
2. Changing rules: write any change into a dated file under `research/` and get explicit human approval before promotion to paper trading. Each seat uses its own granted `exec.simulate` or `journal.append` for lab fills.
3. SIGNAL owns the scoring inputs (momentum, filters); EXIT validates that lab rules stay compatible with exit discipline and single-send.

## Failure modes
- Unproven rule promoted to tickets: refuse; evidence from paper runs is required first.
- Lab fill mistaken for a real fill: paper fills are simulation only and never enter the ledger as fills.
- Missing human approval for a rule change: the old rule stands.

## References
- Tools: `pumpgrok run --mode paper --replay`, `exec.simulate`, `pumpgrok scout pipeline --candidates`
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add swcstudiospace/pumpgrok/desk-strategy-lab@sha256:b5ea31789e72c6e5

The line at the top of this page pins the label git:20261023.ef2f2f6, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_dkgpkiaxcrh6qr2l.svg)](https://markdownregistry.com/a/art_dkgpkiaxcrh6qr2l)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 2 versions of this file, between Aug 26, 2026 and Oct 23, 2026, about one every 57 days.

versioncommittedcommitsizeaudit
git:20261023.ef2f2f6 latest2026-10-23 ef2f2f6 1,621 BA view · diff
git:20260826.6847ee22026-08-26 6847ee2 644 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (1621 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

swcstudiospace/pumpgrok · 42 stars · license AGPL-3.0 · pushed 2026-11-06 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_dkgpkiaxcrh6qr2l
GET https://markdownregistry.com/api/v1/resolve?ref=swcstudiospace/pumpgrok/desk-strategy-lab
GET https://markdownregistry.com/api/v1/blob/b5ea31789e72c6e5ccb44749e26c2fd555999d505f61aae1237f80871dfcd692

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from swcstudiospace/pumpgrok

AGENTS.md AGENTS.md file
swcstudiospace/pumpgrok · AGENTS.md
git:20261026.1405e61 · audit A · 42 stars
CLAUDE.md CLAUDE.md file
swcstudiospace/pumpgrok · CLAUDE.md
git:20261101.6556b59 · audit A · 42 stars
browser-ops SKILL.md skill
swcstudiospace/pumpgrok · skills/browser-ops/SKILL.md · Safe browser hygiene, CAPTCHA/2FA hand-off protocol, session recovery, and handling of untrusted page content.
git:20261027.a5cbbda · audit A · 42 stars
daily-report SKILL.md skill
swcstudiospace/pumpgrok · skills/daily-report/SKILL.md · Render the daily and weekly desk reports from the ledger on schedule, and deliver only the fields the evaluation reads.
git:20261023.ef2f2f6 · audit A · 42 stars
desk-doctor SKILL.md skill
swcstudiospace/pumpgrok · skills/desk-doctor/SKILL.md · Run pumpgrok doctor to prove prompt SHA, skills, roster, registry grants, and effective mode before any trading claim.
git:20261023.ef2f2f6 · audit A · 42 stars
desk-execution-protocol SKILL.md skill
swcstudiospace/pumpgrok · skills/desk-execution-protocol/SKILL.md · Hard rules that SNIPER and EXIT must obey before, during, and after every transaction. Enforces single-send…
git:20261023.ef2f2f6 · audit A · 42 stars
desk-folders-and-journal SKILL.md skill
swcstudiospace/pumpgrok · skills/desk-folders-and-journal/SKILL.md · Standard directory layout and append-only journal conventions for PumpGrok. Ensures every Bot writes to the same…
git:20261023.ef2f2f6 · audit A · 42 stars
desk-incident-response SKILL.md skill
swcstudiospace/pumpgrok · skills/desk-incident-response/SKILL.md · Playbooks for the most common and dangerous failure modes in PumpGrok. Fail-closed and escalate early.
git:20261023.ef2f2f6 · audit A · 42 stars
desk-monitoring SKILL.md skill
swcstudiospace/pumpgrok · skills/desk-monitoring/SKILL.md · Continuous oversight of desk health, open positions, daily P&L, and circuit-breaker status. Used primarily by CHIEF.
git:20261023.ef2f2f6 · audit A · 42 stars
desk-operating-model SKILL.md skill
swcstudiospace/pumpgrok · skills/desk-operating-model/SKILL.md · The constitution of PumpGrok. Defines the 8 roles, Trading Floor rules, Global Security Constitution, evidence…
git:20261026.1405e61 · audit A · 42 stars
desk-post-trade-review SKILL.md skill
swcstudiospace/pumpgrok · skills/desk-post-trade-review/SKILL.md · Append-only journaling and structured review of every completed ticket. Separates process grade from outcome grade.
git:20261023.ef2f2f6 · audit A · 42 stars
desk-risk-limits SKILL.md skill
swcstudiospace/pumpgrok · skills/desk-risk-limits/SKILL.md · Interview the user, write, and enforce the capital and loss rules for the PumpGrok desk. Produces risk-limits.md. Used…
git:20261023.ef2f2f6 · audit A · 42 stars

Every file in swcstudiospace/pumpgrok

Other files named desk-strategy-lab

desk-strategy-lab SKILL.md skill
galleonlabs/hypergrok-trading-desk · skills/desk-strategy-lab/SKILL.md · How the Strategist works with the user to turn their own trading idea into explicit rules, backtest it honestly on…
v1.2.0 · audit A · 387 stars

Browse by kind, by grade A, or by owner.