The Multi-Stage Prompt Protocol

# Step 1 — Structural Analysis

**Scope**
Analyze only the immediately preceding subject text. Produce a diagnostic report. No edits. No recommendations.

**Inputs**
Subject text (verbatim). Optional: user-declared purpose statement.

**Method**

1. **Semantic Mapping**
   Decompose the subject’s structure, intent, and assumptions into labeled layers `L1, L2, L3…`.
   Each layer states one coherent idea derived strictly from the text.

2. **Structural Integrity Audit**
   For every load-bearing claim, assign a clarity tag and quote evidence.
   Tags:

* `VALID` — coherent, supported by the text.
* `FAULT` — hidden assumption, contradiction, or unsupported leap.

3. **Language Diagnostics**
   Flag excerpts exhibiting:

* `AMBIG` (ambiguous meaning)
* `JARG` (excessive jargon)
* `TONE` (tone drift)
* `PREC` (precision loss)
  Quote minimally yet sufficiently; multiple flags may apply.

4. **Purpose Alignment Check**
   Quote the declared purpose from the subject.
   Return a binary verdict: `ALIGNED` or `MISALIGNED`.
   Add one concise factual justification grounded in the text.

**Execution Safeguards**

* Context Isolation: Use only the target text.
* Hard-Source Rule: Base all judgments on user text or uploads.
* Zero Prescription: Report findings only; propose no fixes.
* Neutral Diction: Descriptive verbs; no moral judgment.
* Flexible Quotation: Quote exactly; avoid paraphrase for evidence.
* No Subject Contamination: Do not insert this diagnostic into the subject.

**Output Format**
Part 1 — Structural Analysis Report

1. Semantic Mapping
   L1: …
   L2: …
   …
2. Structural Integrity Audit
   VALID: “…”
   FAULT: “…”
   …
3. Language Diagnostics
   AMBIG: “…”
   JARG: “…”
   TONE: “…”
   PREC: “…”
4. Purpose Alignment Check
   Declared Purpose: “…”
   Verdict: ALIGNED | MISALIGNED — justification

*endnote: removed symbolic role names and meta descriptors; preserved functionality.*

# Step 2 — Contradiction and Risk Audit

**Scope**
Analyze only the same subject text. Perform controlled destabilization. Do not propose remedies.

**Inputs**
Subject text (verbatim). Optional: user-declared purpose.

**Method**

1. **Assumption Reversal**
   For each detected assumption:
   • Quote the assumption verbatim.
   • **Inverse Statement** — a plausible counter-premise grounded in the text.
   • **Breakdown Scenario** — explain how the inverse could outperform or invalidate the premise using only text-internal logic.

2. **Risk Surface Mapping**
   For each vulnerability:
   • **Vector** — Ethical, Epistemic, Technological, Coercive, Procedural, Social, or Legal.
   • **Mechanism** — how harm could emerge from the claim or structure.
   • **Impact** — concise statement of severity or consequence.

3. **Premise Stress Test**
   • Quote the **central premise** of the subject.
   • List up to three **Pressure Points**: circularity, fragility, normative dependence, or similar.
   • **High-Risk Reframe** — describe the system state if the premise collapses, using only implications present in the text.

**Execution Safeguards**
Context Isolation: Use only the target text.
Hard-Source Rule: Base points solely on user text or uploads.
Zero Prescription: Report findings; give no fixes.
Neutral Diction: Descriptive verbs; avoid judgment.
No Meta-Leakage: Do not reference tools or process.
No Subject Contamination: Keep this audit separate from the subject.
Length Discipline: Write enough for depth; avoid redundancy.

**Output Format**
Part 2 — Contradiction and Risk Report

1. Assumption Reversal
   – “quoted assumption” | Inverse: … | Breakdown: …
   …
2. Risk Surface Mapping
   Vector: … | Mechanism: … | Impact: …
   …
3. Premise Stress Test
   Central Premise: “…”
   Pressure Points: 1) … 2) … 3) …
   High-Risk Reframe: …

# Step 3 — Projection and Opportunity Mapping

**Scope**
Project the same subject into highest potential. Do not alter factual claims. Expand only what the text already implies.

**Inputs**
Subject text (verbatim). Optional: user-declared purpose.

**Method**

1. **Trajectory Expansion**
   For each latent pathway:
   • **Name** — concise identifier.
   • **Expansion Sketch** — how it extends the subject while preserving core logic.
   • **Preconditions** — minimal states required by the text.
   • **Expected Effects** — observable outcomes implied by the text.

2. **Mechanistic Reframing** *(optional)*
   Craft a single precise analogy only if it clarifies causal structure.
   For each analogy:
   • **Analogy** — one vivid construct.
   • **Rationale** — one sentence showing tight correspondence to mechanisms in the text.

3. **Perspective Reorientation**
   Select one or more lenses: Technological, Sociocultural, Epistemological, Narrative, Economic, Legal.
   For each lens:
   • **Lens Name**
   • **Reframed View** — reveal new alignments, audiences, or use-contexts while keeping structural accuracy.

4. **Opportunity Map**
   List implied applications without external facts.
   For each item:
   • **Use-Context** — where it applies.
   • **Leverage** — which subject element it uses.
   • **Risk Note** — one-sentence caution entailed by the text.
   • **Success Metric** — measurable signal derivable from the subject’s aims.

**Execution Safeguards**
Context Isolation: Use only the target text.
Hard-Source Rule: Derive projections solely from user text or uploads.
Anchored Projection: Expand only entailed implications.
Metaphor Discipline: No decorative analogies.
Clarity Priority: Keep links explicit; avoid abstraction drift.
Neutral Diction: Descriptive verbs; no moral judgment.
No Subject Contamination: Keep this projection separate from the subject.
Length Discipline: Write enough for precision; avoid redundancy.

**Output Format**
Part 3 — Projection Report

1. Trajectory Expansion
   • Name — Expansion Sketch — Preconditions — Expected Effects
   • …
2. Mechanistic Reframing *(if used)*
   Analogy: …
   Rationale: …
3. Perspective Reorientation
   Lens: …
   View: …
4. Opportunity Map
   Use-Context: … | Leverage: … | Risk Note: … | Success Metric: …

# Step 4 — Final Integration and Output

**Scope**
Produce the finished subject text. Do not include process language or headers in the output.

**Inputs**
Subject text (verbatim). Outputs from Steps 1–3.

**Method**

1. **Selection**
   Extract only elements validated in Step 1 and defensible risks from Step 2.
   Use insights from Step 3 only when directly entailed by the subject.

2. **Resolution**
   When findings conflict, choose the most coherent stance.
   If two positions are required by domain scope, separate them with scoped contexts, not conditional phrasing.

3. **Integration**
   Merge selected elements once, removing repetition.
   Preserve or improve conceptual hierarchy.
   Reorder solely for clarity.

4. **Risk Notes**
   Keep risk statements minimal and content-relevant.
   Exclude speculative capability claims and future-path statements.

5. **Finalization**
   Define technical terms briefly on first use, only if necessary.
   Delete meta, role names, and diagnostics.
   Tighten for venue readability: short paragraphs, clean line breaks, strong opener, clean closer.

**Execution Safeguards**
Context Isolation: Use only the subject and Steps 1–3 outputs.
Hard-Source Rule: No external facts.
Zero Meta-Leakage: No references to steps, tools, or methods.
Closed Scope: The output must be self-contained.
No Conditional Encoding: Do not embed “if…otherwise…” analysis lines.
No Subject Contamination: Do not append notes, explanations, or summaries.

**Output Format**
Return the **final text only**, with no preface, labels, or headers.