cheat-on-content
You're reading this. The skill predicted it. A workflow that turns every post into a calibrated experiment—score, blind-predict, retro, evolve. The future doesn't reward effort, it rewards those who see the pattern first. 1M followers in a month — not luck, system.
Cheat on Content is an agent skill designed for content creators that transforms content publishing into a data-driven feedback loop. It enables AI agents to score drafts, make blind predictions about performance, log decisions, and conduct retrospectives based on actual results. The skill installs 14 sub-skills into compatible agents and maintains an evolving scoring rubric that learns from each piece of content published. Over time, it develops channel-specific judgment by tracking predictions against real outcomes and automatically prompting formula refinements when patterns shift.
Key Features
Use Cases
- 01Content creators tracking which video scripts will perform well before filming or publishing
- 02YouTubers and social media creators building a personalized hit-prediction model over 1-3 months
- 03Content teams conducting systematic post-mortems on every published piece to extract learnable patterns
- 04Solo creators replacing subjective 'gut feel' with logged predictions and data-validated retrospectives
- 05Marketing teams calibrating content scoring rubrics against benchmark competitors in their niche
- 06Multi-platform creators managing a content buffer and prioritizing which pieces to ship based on predicted ROI
Related Skills
View moresuperpowers
An agentic skills framework & software development methodology that works.
skills
Skills for Real Engineers. Straight from my .agents directory.
skills
Public repository for Agent Skills
ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
cheat-on-content — FAQ
What is Cheat on Content?+
Cheat on Content is an agent skill for content creators that turns every post into a calibrated experiment. It enables AI agents to score drafts, make blind predictions about performance, and conduct retrospectives based on actual data, building a channel-specific prediction model over time.
How do I install the Cheat on Content agent skill?+
Clone the GitHub repository, navigate into the directory, and run `bash install.sh` to symlink 14 sub-skills into your agent's skill directory. For Codex agents use `bash install.sh --codex`, or `bash install.sh --all` for both Claude Code and Codex support.
Which AI agents work with this skill?+
The skill supports Claude Code (default) and Codex agents. After installation, initialize it in any content project directory by opening a compatible agent and saying 'init cheat-on-content'.
Do I need API keys or a benchmark account to use this?+
No API keys are required. However, importing a benchmark account (5-10 sample posts from a comparable channel) is strongly recommended to anchor predictions immediately; without one, the first 5 predictions will have approximately ±50% precision.
Is Cheat on Content free to use?+
Yes, it is released under the MIT license, which permits free use, modification, and even commercial or closed-source integration.
How does this differ from using ChatGPT or other general AI assistants?+
General LLMs provide global-average advice and forget context between sessions. This skill builds a channel-specific scoring model from your actual publishing history, remembers every prediction and outcome, and evolves its rubric based on your performance data—it serves only your channel, not everyone's.
How do I install cheat-on-content?+
Open the source repository on GitHub and follow its README. cheat-on-content is a skill — MCP Agents Market links you directly to the official repo.
Is cheat-on-content free?+
cheat-on-content is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.