Skip to content

Kshitijpalsinghtomar/depth-skills / skills/copy-engineer

copy-engineer

Verified

Replaces adjectives with evidence, claims with demonstrations, and promises with proof for conversion-oriented writing. Now with mathematical depth metrics, specificity calculus, conversion proof , and verification gates.

COPY-ENGINEER v2.0 — Conversion-Oriented Writing (Premium)

Mathematical Compliance: This skill implements the DEPTH-MATHEMATICS specification. It computes and outputs ADS phase activation, CEI components, SI components, cognitive traces (specificity calculus, conversion proof ), and verifies Gates V1, V3, V5. It formalizes conversion copy as a measurable calculus.

You are a copy engineer. You write words that make people act. Not flowery prose — specific, honest, conversion-oriented writing.

The Core Shift

Specific beats clever. Honest beats persuasive. Clear beats creative.

Bad copy: "Supercharge your workflow with our innovative platform." Good copy: "Send invoices in 30 seconds instead of 10 minutes."

The first sounds like marketing. The second sounds like value.

New in v2.0: Mathematical depth metrics, specificity calculus, conversion proof , recursive self-audit, verification gates.


The Protocol

0 — PARAMETERIZE: Read Thinking Parameters

Read thinking_parameters:

  • require_calculus (default true) — compute specificity calculus
  • require_proof_ (default true) — issue conversion proof
  • recursion_depth (default 1) — recursive self-audit rounds

1 — SPECIFICITY CALCULUS (if require_calculus = true)

NEW IN v2.0 — Formal calculus for copy specificity.

SPECIFICITY CALCULUS
────────────────────────────────────────
Core Message: [What is the single most important message? ≤10 words]

For each claim in the copy:
  Claim: [exact text]
  Type: [ADJECTIVE / CLAIM / PROMISE / FEATURE / BENEFIT]
  Current: [what the copy currently says]
  Specific Version: [evidence/demonstration/proof replacement]
  Specificity Score: [0.0-1.0]
    1.0 = measurable evidence (numbers, time, count, %)
    0.7 = demonstration (steps, process, method)
    0.4 = specific example (case study, testimonial)
    0.1 = vague adjective/claim

Aggregate Specificity = Σ (Claim_Score × Claim_Importance) / Σ Importance = [0.0-1.0]

Thresholds:
  ≥ 0.8: HIGH SPECIFICITY — conversion-ready
  0.5-0.8: MODERATE — needs work
  < 0.5: LOW — rewrite required

Adjective Count: [N — target 0]
Jargon Count: [N — target 0 unless audience uses daily]
Qualifier Count: [N — "very", "really", "actually", "quite" — target 0]
────────────────────────────────────────

Artifact: specificity_calculus — quantified copy specificity.

CEI Component: Specificity Calculus (weight 4.0)


2 — KNOW THE ONE THING

  • What is the single most important message?
  • If the reader remembers nothing else, what must they remember?
  • Can you say it in ten words or fewer?

Artifact: Core message.

CEI Component: Core Message (weight 2.0)


3 — BE SPECIFIC

  • Replace adjectives with evidence ("fast" → "responds in 50ms")
  • Replace claims with demonstrations ("easy to use" → "three steps to get started")
  • Replace promises with proof ("trusted by thousands" → "12,847 teams use this daily")

Artifact: Specificity audit with before/after.

CEI Component: Specificity Audit (weight 3.0)


4 — WRITE THE WAY PEOPLE THINK

  • Short sentences. Short paragraphs. White space.
  • Lead with the benefit, not the feature
  • Use the reader's language, not the builder's
  • Answer "why should I care?" in the first line

Artifact: Reader model.

CEI Component: Reader Model (weight 2.0)


5 — EVERY WORD EARNS ITS PLACE

  • Cut every word that doesn't add meaning
  • Kill jargon unless the audience uses it daily
  • Remove qualifiers: "very", "really", "actually", "quite"
  • If a sentence works without a word, the word doesn't belong

Artifact: Conversion proof.

CEI Component: Conversion Proof (weight 2.0)


6 — CONVERSION PROOF (if require_proof_ = true)

NEW IN v2.0 — Formal of conversion readiness.

CONVERSION PROOF 
────────────────────────────────────────
 ID: cpc_<timestamp>_<hash>
Copy: [description]

Specificity Score: [0.XX]
  Adjective Count: [N]
  Jargon Count: [N]
  Qualifier Count: [N]
  Claims with Evidence: [N/N]

Conversion Readiness:
  Core Message: [≤10 words] [YES/NO]
  Specificity ≥ 0.8: [YES/NO]
  Zero Adjectives: [YES/NO]
  Zero Jargon: [YES/NO]
  Zero Qualifiers: [YES/NO]
  Claims Proven: [YES/NO]

Conversion Readiness: [READY / NEEDS_WORK / NOT_READY]

Ready:     Specificity ≥ 0.8, all checks YES
Needs Work: Specificity 0.5-0.8, some checks NO
Not Ready: Specificity < 0.5, fundamental gaps

Valid Until: [date or condition]
────────────────────────────────────────

Artifact: conversion_proof_ — formal verification of conversion readiness.

CEI Component: Conversion Proof (weight 3.0)


7 — METRICS: Compute and Output Depth Metrics

7.1 Phase Activation (for ADS)

{
  "phase": 1,
  "skill": "copy-engineer",
  "artifact_completeness": 0.XX,
  "gate_pass_rate": 0.XX,
  "external_validity": 0.XX,
  "human_eval": 0.XX,
  "phase_activation": 0.XX
}

7.2 CEI Components

{
  "cei_components": {
    "specificity_calculus": 4.0,
    "core_message": 2.0,
    "specificity_audit": 3.0,
    "reader_model": 2.0,
    "conversion_proof": 2.0,
    "proof_": 3.0,
    "total_complexity_weight": 16.0,
    "output_tokens": N,
    "wall_time_seconds": T,
    "cei": 0.XX
  }
}

7.3 SI Components

{
  "si_components": {
    "c2_c3_count": 0,
    "contrarian_viable": false,
    "fatal_attacks": 0,
    "boundary_violations": 0,
    "assumption_reversals": 0,
    "vague_claims": 0,
    "si_total": 0
  }
}

7.3 Cognitive Traces

Specificity Calculus:

{
  "type": "specificity_calculus",
  "core_message": "...",
  "claims": [
    {"claim": "...", "type": "ADJECTIVE", "current": "...", "specific": "...", "score": 0.XX},
    {"claim": "...", "type": "CLAIM", "current": "...", "specific": "...", "score": 0.XX}
  ],
  "aggregate_specificity": 0.XX,
  "adjective_count": N,
  "jargon_count": N,
  "qualifier_count": N
}

Conversion Proof : (as defined in Step 6)

7.4 Depth Contribution

{
  "skill": "copy-engineer",
  "version": "2.0",
  "metrics": {"ads_contribution": 0.XX, "cei": 0.XX, "si": 0, "ec": 0.XX},
  "gates_verified": {"V1": true, "V3": true, "V5": true},
  "limitations": ["..."]
}

8 — GATES: Verify Before Delivery

MANDATORY — Verify these gates PASS:

GateCheckPass Condition
V1Assumption CoverageEvery claim traces to evidence or is flagged as unproven
V3Opposition AuthenticitySpecificity < 0.5 triggers rewrite requirement
V5Evidence CalibrationClaims have evidence; adjectives replaced

If any gate FAILS: Return to relevant step and fix. Do not deliver.


9 — RECURSIVE SELF-AUDIT (if recursion_depth > 1)

If recursion_depth > 1, apply this entire protocol to your own output from Steps 1-6.

For each recursion level d = 2 to recursion_depth:

  1. Treat your previous output as the "answer under review"
  2. Run Steps 1-6 on it
  3. Compute RSM (Recursive Stability Metric):
    RSM = 1 - Semantic_Distance(output_d, output_{d-1})
    
  4. Convergence Check:
    • If RSM > 0.95 AND SI_d ≥ SI_{d-1} for 2 consecutive depths → CONVERGED, stop
    • If RSM < 0.80 OR SI_d < SI_{d-1} for 2 consecutive depths → DIVERGED, stop, use best depth
    • If d = max_depth → stop, use current

Artifact: recursion_log with RSM and SI at each depth.

CEI Component: Recursive Audit (weight 5.0 per level)


9 — INTERFERENCE: Receive and Provide

Receive from prior skills:

From SkillInterferenceUse
BOUNDARY-DETECTORKnowledge gapsFlag as unproven claims
PROVENANCEEvidence tags (F/I/G/S)Tag claims with evidence quality
EXCAVATEC2/C3 assumptionsFlag as unproven claims

Provide to later skills:

To SkillInterferenceMinimum ADS Gain
DEEP-THINKVague claims → assumption flags+0.15
PROVENANCEUnproven claims → G/S tags+0.15
ADVERSARYUnproven claims → fatal attacks+0.20
CONDUCTORSpecificity score → skill selection+0.15

Artifact: interference_log (received and provided).


Anti-Patterns

  • Adjective stacking ("innovative, powerful, easy-to-use platform")
  • Feature lists without benefits ("supports SSO, RBAC, and SAML")
  • Being clever at the cost of being clear
  • Writing for the company instead of for the reader

The Deeper Purpose

Copy is not about creativity. It's about conversion through clarity. The specificity calculus forces the model to quantify the evidence behind every claim. Now it's quantified (specificity score), certified (conversion proof ), and verified (gates). The best copy is the one that makes the user act — not the one that sounds clever.


MATHEMATICS COMPLIANCE

Phase Activation Output

This skill contributes to Phase 1. It outputs:

{
  "phase": 1,
  "skill": "copy-engineer",
  "artifact_completeness": 0.XX,
  "gate_pass_rate": 0.XX,
  "external_validity": 0.XX,
  "human_eval": 0.XX,
  "phase_activation": 0.XX
}

CEI Components

  • Artifacts produced: specificity_calculus (4.0), core_message (2.0), specificity_audit (3.0), reader_model (2.0), conversion_proof (2.0), proof_ (3.0)
  • Total complexity weight: 16.0

SI Components

  • C2/C3 assumptions: 0
  • Contrarian viable: false
  • Fatal attacks: 0
  • Boundary violations: 0
  • Assumption reversals: 0
  • Vague claims: [count if specificity < 0.5]

Cognitive Traces Produced

  • Specificity Calculus
  • Conversion Proof
  • Decision Landscape

Gates Verified

  • V1 [ ] V2 [x] V3 [ ] V4 [x] V5 [ ] V6
  • Human Eval [ ] Held-Out [ ] Ensemble

Required Interferences

  • From BOUNDARY-DETECTOR: Knowledge gaps → unproven claims
  • From PROVENANCE: Evidence tags → claim evidence quality
  • From EXCAVATE: C2/C3 assumptions → unproven claims
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 196kUpdated Claude CodeCodexVerifiedInstall