An agent skill for finding promising X/Twitter conversations and drafting replies people have a reason to answer or share.
The workflow starts with the actual comments: read the post, identify what earns attention and what is missing, then develop a distinct contribution in your voice. It supports learning, drafting, and posting text you have approved.
- Choose conversations using audience fit, visible activity, and room for a useful contribution.
- Study why an actual reply earns attention, while accounting for timing and account advantages.
- Explore different angles privately and choose one that adds more than agreement or praise.
- Check for a reason to answer, remember, or share the reply, without copying the leading comment.
- Adapt language, tone, and topic to each user. A precise professional question can be a better contribution than a joke.
- Learn from real results when available, keeping reach separate from meaningful interaction.
These are ways to improve opportunity selection and draft quality, not a promise of virality. Research and limitations distinguish X's documented behavior from broader research and editorial judgment. Evaluation cases check the skill's decisions across topics and languages; they do not measure real-world reach.
Clone the repository into your agent's skills directory. For a local Codex installation using ~/.codex/skills:
git clone https://github.com/nniiovoo/cheat-reply.git ~/.codex/skills/cheat-replyFor Claude Code:
git clone https://github.com/nniiovoo/cheat-reply.git ~/.claude/skills/cheat-replyIf cheat-reply is already installed at the destination, compare or back up your customized copy before replacing it. Other skill-compatible agents can load the root SKILL.md; use their supported skill directory.
The package contains instructions and examples. It does not install a browser extension, connect an X account, store credentials, or run a background posting service. Live research needs browsing tools that can read the conversation; you can also paste the post and relevant comments.
Example requests:
Use $cheat-reply to study the replies to this post and draft one response
in my voice: [post URL]. Aim for relevant reach, explain what the reply
adds, and keep it draft-only.
Use $cheat-reply to find active conversations about [your topic].
Study the comments and show me three relevant targets. Do not post.
Use $cheat-reply to learn from these pasted comments. Summarize the
patterns worth testing; don't write a reply yet.
Research and drafting do not authorize publication. Posting requires approval of the exact reply and target, an authorized posting tool or signed-in browser, and verification after sending.
Describe your audience and voice in the request, or copy the optional templates into your own working project:
- audience.md: audience, voice, and boundaries.
- reply-learnings.md: current observations and recent cases.
The skill reads those files only when present. It does not require the wider cheat-on-content system. Keep real account notes, private conversations, credentials, and performance data outside this repository.
| File | Purpose |
|---|---|
| SKILL.md | Main workflow and posting boundaries |
| Reply playbook | Optional strategy and maintenance reference |
| Research and limitations | Primary sources and boundaries of the evidence |
| Evaluation cases | Reproducible dry-run scenarios |
| Examples | Generic, optional account-context templates |
| Agent metadata | Skill discovery metadata |
Issues and pull requests are welcome. Include a concrete example of the behavior you want to improve. Anonymize account details and private context; do not submit credentials or copied private conversations. Keep new rules specific to the situation where they help, and distinguish observations from claims about the algorithm.
This standalone distribution is maintained by nniiovoo. It was developed from a local extension to cheat-on-content, with additional reply research and workflow refinements.
Distributed under the MIT License, retaining the cheat-on-content contributors notice for inherited material. Public packaging adds generic context templates and removes account-specific case histories and local-path dependencies.