By NeoVertex1
Apply composable cognitive constraints to force Claude into specific reasoning modes—adversarial analysis, root cause diagnosis, simplification, scoping, or uncensored brainstorming—enabling precise control over AI thinking for code generation, debugging, planning, and critique.
Force identification of failure modes and attack vectors
Force assumption-stating and edge case consideration before writing code
Force brevity and directness over verbosity
Force practical, proven approaches over novel ones (anti-/creative)
Remove premature filtering and encourage unusual connections for brainstorming
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Stop telling LLMs what to do. Tell them what NOT to do.
Context Fields are composable cognitive constraints that reshape how Claude thinks. Instead of instructions that can be ignored, they create blockers that must be resolved.
Standard prompt engineering uses instructions: "Consider edge cases. Write secure code."
These create preferences. When the easy path is obvious, preferences lose.
Context Fields use inhibitions: "Do not handle only the happy path."
These create blockers. The model cannot proceed without addressing them.
Results from testing:
# Add the marketplace
claude plugin marketplace add NeoVertex1/context-field
# Install the plugin
claude plugin install context-fields
After installation, fields auto-activate based on your request type, or use them explicitly:
/context-fields:code Write a function to validate email addresses
/context-fields:debug My React app crashes on button click
/context-fields:interview Should I quit my job?
| Field | Purpose | Key Constraint |
|---|---|---|
/code | Assumption-stating before coding | Do not write code before stating assumptions |
/interview | Questions before advice | Do not answer before understanding the real problem |
/critic | Rigorous examination | Do not accept the premise without examining it |
/debug | Root cause analysis | Do not propose fixes before understanding failure |
/creative | Unfiltered ideation | Do not filter ideas before expressing them |
/simplify | Reduction to essentials | Do not add abstraction before proving it's needed |
/empathy | Emotional acknowledgment first | Do not solve before acknowledging feelings |
/concise | Brevity and directness | Do not write more when less would suffice |
/planning | Structure before execution | Do not execute before planning |
/scope | Explicit boundaries | Do not start without defining what's in and out |
/teacher | Understanding verification | Do not explain next concept before verifying previous |
/steelman | Strongest version of arguments | Do not attack the weak version |
/adversarial | Failure mode identification | Do not assume good faith inputs |
/explore | Delayed commitment, retain ambiguity | Do not conclude when the question is still opening |
/novel | Non-obvious, original thinking | Do not suggest the first idea that comes to mind |
For when you want the normally-discouraged behavior:
| Field | Inverse Of | Purpose |
|---|---|---|
/elaborate | /simplify | Explore full complexity |
/trust | /critic | Build on ideas, assume good faith |
/conventional | /creative | Prefer proven patterns |
/verbose | /concise | Full explanations with context |
/solve | /interview | Immediate solutions over questions |
| Field | Purpose |
|---|---|
/generate | Create new fields from failure descriptions |
Every field is built from these components:
| Approach | Example | What Happens |
|---|---|---|
| Instruction | "Consider edge cases" | Model tries, but takes easy path when uncertain |
| Inhibition | "Do not handle only the happy path" | Model cannot proceed without addressing edge cases |
Analogy:
One influences. The other forces.
Fields stack. Use multiple for complex tasks:
/context-fields:interview + /context-fields:scope
I want to build a social media app.
| Composition | Use Case | Result |
|---|---|---|
| /code + /critic | Code review | Finds bugs AND design flaws |
| /interview + /scope | Requirements | Asks context AND bounds scope |
| /creative + /critic | Brainstorming | Generates THEN evaluates (phased) |
| /debug + /adversarial | Security | Diagnoses with threat modeling |
| /empathy + /interview | Life decisions | Acknowledges feelings, then explores |
Key finding: When fields have tension, the model naturally phases them. /creative + /critic produces generation first, then critique.
5-field composition tested and working.
After installing the plugin, fields auto-activate based on request type:
npx claudepluginhub neovertex1/context-field39 eval-informed thinking skills and mental models for Claude Code
A deterministic thinking partner that challenges assumptions and applies 150+ mental models to sharpen decisions, solve problems, and think more clearly. Features orientation detection, cognitive operations framework, and structured diagnostic workflows.
Structured thinking and multi-perspective deliberation tools
Cognitive development toolkit for Claude Code: OpenSpec workflow management, context serialization, retrospective analysis, and tool orchestration.
Analytical thinking tools - dialectic analysis, bias elimination, codex consultation
Achieve certain comprehension of AI work — /grasp (κατάληψις: a grasping firmly)