MindFluence plugin
MindFluence teaches a model to write persuasive copy using 20 cognitive biases and checks the draft against 13 anti-patterns.
- 77 ★ on the repository
- Last commit
Minimal access
Why this level
- The skill only sets text instructions, but persuasion techniques can mislead readers if the data is wrong.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/mindfluenceDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skill only sets text instructions, but persuasion techniques can mislead readers if the data is wrong.
This is third-party code. Review the repository files before installing.
What it does
MindFluence is a skill prompt for marketing copy. It turns the model into a strategist who picks a set of 20 cognitive biases for the audience, product and platform: social proof, anchoring, framing, loss aversion and others. Each bias in the text is labeled and each decision explained. A 38 row router table makes the choice: audience temperature, product type and platform give a ready set of 3–5 biases. There are 14 prebuilt combinations such as the Trust Spiral and the Urgency Engine, seven tones from `bold-sell` to `luxe-minimal` and four modes: quick, deep, audit and optimize. Before delivery the skill checks the text against 13 anti-patterns, including fake scarcity and fear without an exit, and a seven-point verification.
Who it is for. Marketers and product owners who write landing pages, emails and posts and want to understand which principle supports every phrase.
Good fit when
- A landing page is in preparation and a cold audience needs a set of techniques with reasoning.
- Finished ad copy does not convert and audit mode looks for anti-patterns in it.
- A product launch needs the Launch Day Stack with honest urgency.
- The same idea must be delivered in the `expert-calm` and the `warm-human` tone.
- The output of an SEO skill needs persuasive copy without losing keywords.
Not a fit when
- You need health or finance copy with guaranteed results: the skill does not check legal requirements.
- You want to push fear without a solution: the anti-patterns ban such copy.
- You need a neutral reference description with no persuasion techniques.
Example request
Write a LinkedIn post about launching a service for freelancers in Quick mode with the `expert-calm` tone and label each cognitive bias.Limitations
The skill remains an instruction for the model and does not verify facts: proof numbers, reviews and sources come from the user, otherwise the vague social proof anti-pattern fires. Persuasion techniques act on people, so check the text against advertising law and brand ethics. Results depend on the model: the README recommends GPT, Claude, Gemini, DeepSeek or another capable model. Integration with RankWise and human-ai needs those projects installed separately. The installer writes files to Claude Code, OpenCode, Codex and `~/.agents/skills` folders.
How to disable. Delete the `~/.claude/skills/mindfluence` folder and the same folders for OpenCode, Codex and `~/.agents/skills`. A plugin is removed with `/plugin uninstall`.
FAQ
How do I install MindFluence in Claude Code?
Run `python install.py --target claude` in the clone or copy `skills/mindfluence/` to `~/.claude/skills/mindfluence/`. The `--target all` option installs to Claude Code, OpenCode, Codex and the shared `~/.agents/skills` folder.
Which modes exist?
Quick by default, deep with five clarifying questions at once, audit of finished copy against 13 anti-patterns and optimize for weak copy with A/B variants and confidence thresholds.
Which tones are supported?
Seven: `bold-sell`, `expert-calm`, `rebel-edgy`, `warm-human`, `luxe-minimal`, `community-build` and `data-vivid`. A hybrid is allowed: 70 percent primary cadence and 30 percent lexical markers of a second.
What is the bias router?
A single 38 row table in `SKILL.md`: audience, product and platform give a set of 3–5 biases at once. Scenarios like a launch or an apology swap the set from a quick reference table.
How is the text verified before delivery?
The skill runs 13 mechanical anti-pattern checks, for example searching for words like thousands near customer claims, then a seven-point verification. Fake scarcity without a reason counts as a failure.
Discussion
No questions or reviews yet.
Related
PluginsTwo skills for English prose: humanize rewrites a model draft in a natural voice, and ai-check flags machine-writing tells
One skill with 14 modules for a marketer: a 0–100 site audit, copy, hooks, ads, GEO, email and launches
Slop Stop
Стоп-Штамп (Slop Stop)
Russian-language skill that strips typical AI writing patterns: empty openers, template structures and monotonous rhythm
A skill that removes AI tone from work writing while keeping every fact, commitment and caveat at the same strength.
A plugin of 14 marketing skills: rivals, post calendars, SEO briefs, ads, email, video and approval-gated publishing
Marketing Skills for AI agents
Marketing Skills for AI Agents
About fifty skills for CRO, copywriting, SEO, analytics, ads and launches, built for coding agents
README in short
The skill is self-contained: the core sits in one `SKILL.md` and references load on demand. The documentation is bilingual, in English and Russian.
README badge
Original SKILL.md
--- name: mindfluence description: "Use when creating, auditing, optimizing, or rewriting marketing copy: posts, ads, landing pages, emails, webinars, push notifications, product launches. Applies 20 cognitive biases from behavioral economics and evolutionary psychology to engineer persuasion that converts without inventing facts. Self-contained core: a 38-row bias router, 14 power combos, 13 anti-patterns with detection rules, 7 tone styles, cultural adaptation for 12 regions, and a mandatory PROVENANCE/TRUTH GATE. Use ONLY for marketing content work. DO NOT use for general writing, technical documentation, SEO structure (RankWise), angles and hooks (AngleCraft), or humanizing AI text (human-ai)." license: MIT compatibility: Plain Markdown. No external dependencies. metadata: author: Yevhen Leonidov version: "2.5.1" biases: 20 languages: [any] tone_styles: 7 scenarios: 13 modes: 4 regions: 12 power_combos: 14 anti_patterns: 13 router_rows: 38 narrative_modes: 2 verification_checks: 7 repository: "https://github.com/MADEVAL/MindFluence" --- # MindFluence - Cognitive Bias Marketing Engine > **Tagline:** Engineer persuasion by understanding the brain, not manipulating it. > **Version:** 2.5.1 > **Mode:** Hybrid - fast generation by default; deep customization, audit, and metric-based optimization on request. > **Architecture:** Self-contained core. All tables, anti-patterns, and cultural data needed for standard tasks are inlined. Scenario files are optional depth for complex tasks - see the External Reference Files section below. > **Language-agnostic:** Generates content in any language. Adapts cultural references to the target locale. --- ## Bias Quick-Reference Card ``` #1 SocialProof [SOC] #11 FundAttrErr [SOC] #2 Anchoring [OPT] #12 SunkCost [OPT] #3 Framing [FIL] #13 StatusQuo [FIL] #4 Authority [SOC] #14 FalseConsensus [SOC] #5 Fear/LossAversion [FIL+SOC] #15 InGroup [SOC] #6 Availability [OPT] #16 HaloEffect [OPT] #7 Confirmation [FIL] #17 HindsightBias [OPT] #8 CogDissonance [FIL+OPT] #18 BackfireEffect [FIL] #9 Survivorship [OPT] #19 BiasBlindSpot [FIL] #10 Endowment [OPT] #20 GroupPolarization [SOC]