cheatcode_
Demo data

Distribution infrastructure for proven ideas

cheatcode_

The newest AI technology, deployed in one click.

The ladder

Top ten by weekly runs, metered server side. Momentum compares this week with last. Full ladder, 30 listings

  1. 1standup-synthesizerCommunity sample · demo data

    Turns yesterday's commits, tickets and threads into a 90-second standup brief per teammate.

    Operations3,667runs / wkdown 15%
  2. 2rfp-draftsmanCommunity sample · demo data

    Drafts RFP responses from your past wins, flagging every claim it could not source.

    Sales2,686runs / wkdown 2%
  3. 3PersonaLensCharles Ji · first party · demo data

    Persona agents grown from client records preview customer reaction to every strategic move before you commit.

    Strategy1,282runs / wkdown 2%
  4. 4churn-autopsyCommunity sample · demo data

    Reconstructs why an account really left from tickets, usage logs and the goodbye email.

    Research1,109runs / wkdown 2%
  5. 5compliance-redteamCommunity sample · demo data

    Attacks your agent configs with 200 seeded prompt-injection cases before your users do.

    Engineering1,007runs / wkup 7%
  6. 6inboxforgeCharles Ji · demo data

    Tracks mutual commitments across email threads and generates a prioritized reply queue in your voice.

    Operations990runs / wkup 2%
  7. 7Analytics with Claude CodeadityawrkGitHub ↗

    Metric reconciliation and calculation, A/B readout, data quality, EDA, explain and optimize SQL, weekly report, systematic debugging.

    Analytics0runs / wkno runs yet
  8. 8Anti-Sycophantmachinesoul11GitHub ↗

    Three skills that stop an AI assistant from agreeing with a bad product idea: prove the premise, hobby or business, one real conversation.

    Validation0runs / wkno runs yet
  9. 9Asgard Analytics Subsetasgard-ai-platformGitHub ↗

    Statistics, forecasting and e-commerce analytics: EDA, hypothesis tests, A/B, causal inference, cohorts, dashboards, SQL optimization, ARIMA and Prophet forecasts, RFM.

    Analytics0runs / wkno runs yet
  10. 10ASO SkillsappeekyGitHub ↗

    40 App Store Optimization skills: keyword research, metadata, competitor tracking, screenshots, reviews, and launch plans for iOS and Android.

    Competitive0runs / wkno runs yet
Rank moves only through install and run events. Listings marked demo data carry seeded counts.See all 30 listings

Start from your role

A stage by stage workflow per role, each stage mapped to the skills that do it.

First listing · published first party

PersonaLens

Persona agents grown from client records preview customer reaction to every strategic move before you commit.

Directions testedengine: sampleDemo data
  1. #09Annual-only pricing for SMB tier7.6 of 10
  2. #07Free migration service, 90 days7.4 of 10
  3. #04Usage-based billing pilot6.1 of 10
  4. #11Raise base price 15%5.5 of 10
  5. #02Sunset the legacy plan4.7 of 10
  6. #12Bundle AI onboarding as a paid add-on ($29/mo) instead of raising the base price4.4 of 10
  7. #05Paywall the reporting module4.2 of 10

Weighted satisfaction across 4 customer segments, one persona agent per segment, weighted by its share of client records.

How it works

Three steps. The third one is the whole point.

1

Publish once

One package in the Agent Plugins layout: a manifest, the skills, a license. Indexed open source repositories are packaged at a pinned commit under their own license, and their authors can claim or remove the listing at any time.

2

Install in one click

One command puts the package where Claude Code, ChatGPT Codex, Cursor, Kimi and the rest already look. The archive's sha256 is checked before any file is written.

3

Usage decides the rank

Runs are metered server side. Weekly runs, retention and momentum are computed from event rows and never stored as numbers anyone can edit.

For builders

Publish once, run everywhere

A listing is a directory. The manifest names the capability, the skills say what it does, the license says on what terms. Cheatcode stores and serves the package; the target workspace runs it. Records stay in the user's workspace.

Publisher accounts are not open yet. Until they are, the ladder is seeded from public repositories under permissive licenses, shown with their author and source, never presented as ours.

# one package, every client
plugins/personalens/
├── .claude-plugin/plugin.json   name, version, author
├── cheatcode.json               listing metadata
├── skills/personalens/SKILL.md  what it does, how it runs
├── README.md
└── LICENSE

$ curl -fsSL https://api.cheatcodehq.com/install.sh | sh -s -- personalens --to all
installs into every supported client