diff --git a/skills/competitive-analysis/SKILL.md b/skills/competitive-analysis/SKILL.md
index 825655b..cab96f2 100644
--- a/skills/competitive-analysis/SKILL.md
+++ b/skills/competitive-analysis/SKILL.md
@@ -50,9 +50,9 @@ template and a required owner. We charge per workspace, not per seat.
 Focus areas: pricing model, API/integration depth, permissions model, mobile experience
 ```
 
-If no dimensions are specified, the skill defaults to the five standard dimensions (see Phase 1).
+If no dimensions are specified, the skill defaults to the five standard dimensions (see Step 1).
 
-## Phase 0: Scan Existing Artefacts (Automatic)
+## Step 0: Scan Existing Artefacts (Automatic)
 
 Before scoping the analysis, silently scan for existing strategy and competitive analysis documents:
 
@@ -65,11 +65,11 @@ find . -maxdepth 3 -name "competitive-analysis.md" 2>/dev/null | head -3 | xargs
 Based on what you find:
 - **If COMPETITIVE-ANALYSIS.md exists**: summarise what's already documented and ask whether to update the existing analysis or start fresh. If updating, carry forward existing players and dimensions as the baseline.
 - **If STRATEGY.md exists**: extract the strategic priorities that inform which competitive dimensions matter most for this context.
-- **If nothing is found**: proceed to Phase 1 as normal.
+- **If nothing is found**: proceed to Step 1 as normal.
 
 Do not mention the scan to the user.
 
-## Phase 1: Scope
+## Step 1: Scope
 
 Before analysing competitors, the skill confirms which dimensions matter most for your context. It will ask:
 
@@ -98,7 +98,7 @@ Before analysing competitors, the skill confirms which dimensions matter most fo
 
 If the user provides evaluation dimensions, those replace or extend the defaults. The analysis always evaluates every competitor on the same dimensions so the comparison matrix is valid.
 
-## Phase 2: Analysis
+## Step 2: Analysis
 
 For each competitor listed, the skill builds a structured profile.
 
@@ -112,7 +112,7 @@ Each profile covers:
 
 ### Capability Assessment
 
-For each dimension selected in Phase 1, the skill assesses each competitor on a three-point scale:
+For each dimension selected in Step 1, the skill assesses each competitor on a three-point scale:
 
 | Rating | Meaning |
 |---|---|
@@ -129,7 +129,7 @@ After the dimension assessment, each profile closes with:
 - **Top 2–3 strengths** — what this competitor does well enough to win deals or retain users
 - **Top 2–3 weaknesses** — where this competitor is most vulnerable or where customers most often complain
 
-## Phase 3: Synthesis
+## Step 3: Synthesis
 
 Once all competitor profiles are complete, the skill synthesises across the full landscape.
 
diff --git a/skills/prioritize/SKILL.md b/skills/prioritize/SKILL.md
index aa43ea0..3db2235 100644
--- a/skills/prioritize/SKILL.md
+++ b/skills/prioritize/SKILL.md
@@ -11,7 +11,7 @@ allowed-tools: [Read, Write, Bash]
 
 Score and rank features or initiatives so the most valuable, strategic work rises to the top. The default framework is RICE — opinionated enough to force a decision, flexible enough to reflect your strategy.
 
-## Phase 0: Context Scan (Silent)
+## Step 0: Context Scan (Silent)
 
 Before asking anything, scan for strategic context:
 
@@ -31,7 +31,7 @@ Use what you find to:
 
 Do not mention the scan.
 
-## Phase 1: Input
+## Step 1: Input
 
 Accept items in any format — a numbered list, bullet points, ticket IDs, FR filenames, or a paragraph. Extract the item name and any context provided.
 
@@ -41,7 +41,7 @@ If the input is sparse (item name only, no context), ask for the missing inputs
 
 **If the user mentions agentic workflows, AI agents, or autonomous execution**, suggest Agentic RICE and explain the difference before proceeding.
 
-## Phase 2: Score Each Item
+## Step 2: Score Each Item
 
 Default framework is **RICE**. Score every item, then rank by RICE score descending.
 
@@ -97,7 +97,7 @@ Derive a touchpoint count from the answers. If all four are clean, Agentic Effor
 - Scope isolation and verifiability: ask the user — these require codebase knowledge the skill doesn't have.
 - Default to Agentic Effort = 1 if uncertain.
 
-## Phase 3: Output
+## Step 3: Output
 
 Save the result as `prioritization-[YYYY-MM-DD].md` in the current directory. Present the ranked table inline and confirm the file was saved.
 
diff --git a/skills/write-product-strategy/SKILL.md b/skills/write-product-strategy/SKILL.md
index 74b4a98..ae30b1c 100644
--- a/skills/write-product-strategy/SKILL.md
+++ b/skills/write-product-strategy/SKILL.md
@@ -15,7 +15,7 @@ This skill helps you build a rigorous, comprehensive product strategy document (
 
 > **Before starting:** If you haven't done a structured brainstorm yet, `/superpowers:brainstorming` can map the problem space before we commit to strategic direction.
 
-### Phase 1: Context Gathering (Automatic)
+### Step 1: Context Gathering (Automatic)
 
 Silently scan the project directory using Read and Bash:
 
@@ -41,7 +41,7 @@ Based on what you find:
 
 Do not mention the scan to the user.
 
-### Phase 2: Input & Validation (Prompt-Responsive)
+### Step 2: Input & Validation (Prompt-Responsive)
 
 If you provide market/user/competitive context upfront, I'll use it directly. If your input is sparse or vague, I'll guide you through key questions section-by-section.
 
@@ -53,7 +53,7 @@ If you provide market/user/competitive context upfront, I'll use it directly. If
 
 If you can't answer these yet, that's OK — we'll work through them together. But a strategy without these is incomplete.
 
-### Phase 3: Generate STRATEGY.md with Research Citations
+### Step 3: Generate STRATEGY.md with Research Citations
 
 I'll structure your input into the standard 7-section format, making transparent **how research findings drove each strategic choice**. This means:
 - Research evidence is explicitly connected to strategic pillars