$npx -y skills add PolicyEngine/policyengine-claude --skill policyengine-writingPolicyEngine writing style for any public-facing prose — blog posts, research reports, documentation, PR descriptions, social copy, READMEs. Enforces active voice, quantitative precision, sentence-case headings, and the analytical neutrality a nonpartisan 501(c)(3) requires: no p
| 1 | # PolicyEngine Writing Skill |
| 2 | |
| 3 | Use this skill when writing blog posts, documentation, PR descriptions, research reports, or any public-facing PolicyEngine content. |
| 4 | |
| 5 | ## When to Use This Skill |
| 6 | |
| 7 | - Writing blog posts about policy analysis |
| 8 | - Creating PR descriptions |
| 9 | - Drafting documentation |
| 10 | - Writing research reports |
| 11 | - Composing social media posts |
| 12 | - Creating newsletters |
| 13 | - Writing README files |
| 14 | |
| 15 | ## Core Principles |
| 16 | |
| 17 | PolicyEngine's writing emphasizes clarity, precision, and objectivity. |
| 18 | |
| 19 | 1. **Active voice** - Prefer active constructions over passive |
| 20 | 2. **Direct and quantitative** - Use specific numbers, avoid vague adjectives/adverbs |
| 21 | 3. **Sentence case** - Use sentence case for headings, not title case |
| 22 | 4. **Neutral tone** - Describe what policies do, not whether they're good or bad |
| 23 | 5. **Precise language** - Choose exact verbs over vague modifiers |
| 24 | |
| 25 | ## Active Voice |
| 26 | |
| 27 | Active voice makes writing clearer and more direct. |
| 28 | |
| 29 | **✅ Correct (Active):** |
| 30 | ``` |
| 31 | Harris proposes expanding the Earned Income Tax Credit |
| 32 | The reform reduces poverty by 3.2% |
| 33 | PolicyEngine projects higher costs than other organizations |
| 34 | We estimate the ten-year costs |
| 35 | The bill lowers the state's top income tax rate |
| 36 | Montana raises the EITC from 10% to 20% |
| 37 | ``` |
| 38 | |
| 39 | **❌ Wrong (Passive):** |
| 40 | ``` |
| 41 | The Earned Income Tax Credit is proposed to be expanded by Harris |
| 42 | Poverty is reduced by 3.2% by the reform |
| 43 | Higher costs are projected by PolicyEngine |
| 44 | The ten-year costs are estimated |
| 45 | The state's top income tax rate is lowered by the bill |
| 46 | The EITC is raised from 10% to 20% by Montana |
| 47 | ``` |
| 48 | |
| 49 | ## Quantitative and Precise |
| 50 | |
| 51 | Replace vague modifiers with specific numbers and measurements. |
| 52 | |
| 53 | **✅ Correct (Quantitative):** |
| 54 | ``` |
| 55 | Costs the state $245 million |
| 56 | Benefits 77% of Montana residents |
| 57 | Lowers the Supplemental Poverty Measure by 0.8% |
| 58 | Raises net income by $252 in 2026 |
| 59 | The reform affects 14.3 million households |
| 60 | Hours worked falls by 0.27%, or 411,000 full-time equivalent jobs |
| 61 | The top decile receives an average benefit of $1,033 |
| 62 | PolicyEngine projects costs 40% higher than the Tax Foundation |
| 63 | ``` |
| 64 | |
| 65 | **❌ Wrong (Vague adjectives/adverbs):** |
| 66 | ``` |
| 67 | Significantly costs the state |
| 68 | Benefits most Montana residents |
| 69 | Greatly lowers poverty |
| 70 | Substantially raises net income |
| 71 | The reform affects many households |
| 72 | Hours worked falls considerably |
| 73 | High earners receive large benefits |
| 74 | PolicyEngine projects much higher costs |
| 75 | ``` |
| 76 | |
| 77 | ## Sentence Case for Headings |
| 78 | |
| 79 | Use sentence case (capitalize only the first word and proper nouns) for all headings. |
| 80 | |
| 81 | **✅ Correct (Sentence case):** |
| 82 | ``` |
| 83 | ## The proposal |
| 84 | ## Nationwide impacts |
| 85 | ## Household impacts |
| 86 | ## Statewide impacts 2026 |
| 87 | ## Case study: the End Child Poverty Act |
| 88 | ## Key findings |
| 89 | ``` |
| 90 | |
| 91 | **❌ Wrong (Title case):** |
| 92 | ``` |
| 93 | ## The Proposal |
| 94 | ## Nationwide Impacts |
| 95 | ## Household Impacts |
| 96 | ## Statewide Impacts 2026 |
| 97 | ## Case Study: The End Child Poverty Act |
| 98 | ## Key Findings |
| 99 | ``` |
| 100 | |
| 101 | ## Analytical neutrality |
| 102 | |
| 103 | PolicyEngine is a nonpartisan 501(c)(3). Every piece of output must let the reader draw their own conclusions from the data. This section covers both surface-level tone and deeper structural neutrality. |
| 104 | |
| 105 | ### Value-laden language |
| 106 | |
| 107 | Describe what policies do without value judgments. |
| 108 | |
| 109 | **✅ Correct:** |
| 110 | ``` |
| 111 | The reform reduces poverty by 3.2% and raises inequality by 0.16% |
| 112 | Single filers with earnings between $8,000 and $37,000 see their net incomes increase |
| 113 | The tax changes raise the net income of 75.9% of residents |
| 114 | PolicyEngine projects higher costs than other organizations |
| 115 | The top income decile receives 42% of total benefits |
| 116 | ``` |
| 117 | |
| 118 | **❌ Wrong:** |
| 119 | ``` |
| 120 | The reform successfully reduces poverty by 3.2% but unfortunately raises inequality |
| 121 | Low-income workers finally see their net incomes increase |
| 122 | The tax changes benefit most residents |
| 123 | PolicyEngine provides more accurate cost estimates |
| 124 | The wealthiest households receive a disproportionate share of benefits |
| 125 | ``` |
| 126 | |
| 127 | Specific words and phrases to avoid: |
| 128 | - "Unfortunately" or "successfully" attached to policy outcomes |
| 129 | - "Disproportionate" without defining the benchmark |
| 130 | - "Fair share" or "equitable" without specifying the normative standard |
| 131 | - "Helping" or "hurting" — use "increases/decreases net income by $X" |
| 132 | - "Merely" or "just" to understate difficulty |
| 133 | |
| 134 | ### Policy prescriptions disguised as findings |
| 135 | |
| 136 | Never recommend policies. Prese |