FN-7499: require before-after task summaries

Require generated task specs to summarize the requested before-to-after transformation near the top.

- Add a Before → After Transformation section to standard and fast planning prompt templates.\n- Document the new task definition section and cover it with prompt regression tests.\n- Add a patch changeset for the published Fusion package.\n\nFiles changed:\n .changeset/fn-7499-before-after-transformation.md |  7 +++++++\n docs/task-management.md                           |  1 +\n packages/core/src/__tests__/agent-prompts.test.ts | 17 +++++++++++++++\n packages/core/src/agent-prompts.ts                | 25 +++++++++++++++++++----\n packages/engine/src/__tests__/triage.test.ts      | 20 ++++++++++++++++++\n 5 files changed, 66 insertions(+), 4 deletions(-)

Fusion-Task-Id: FN-7499

Fusion-Task-Lineage: d049b5d6-a4bc-40dd-831d-04a11f9dc2cf

Co-authored-by: Fusion (runfusion.ai) <noreply@runfusion.ai>

.changeset/fn-7499-before-after-transformation.md

@@ -0,0 +1,7 @@
+---
+"@runfusion/fusion": patch
+---
+
+summary: Add before-to-after transformation summaries to generated task definitions.
+category: feature
+dev: Built-in standard and fast triage prompts now require a `## Before → After Transformation` section.

docs/task-management.md

@@ -543,6 +543,7 @@ The task detail modal exposes multiple tabs:
 After planning, each task gets a structured `PROMPT.md` with sections like:
 
 - Mission
+- Before → after transformation summary
 - Dependencies
 - Context to read first
 - File scope

packages/core/src/__tests__/agent-prompts.test.ts

@@ -295,6 +295,23 @@ describe("resolveAgentPrompt", () => {
     expect(fastPrompt.split("\n").length).toBeLessThan(120);
   });
 
+  it("requires before-to-after transformation summaries across planning prompts", () => {
+    const prompts = [
+      resolveAgentPrompt("triage"),
+      builtinSeamPrompt("planning"),
+      builtinSeamPrompt("planning-fast"),
+    ];
+
+    for (const prompt of prompts) {
+      expect(prompt).toContain("## Before → After Transformation");
+      expect(prompt).toContain("Before");
+      expect(prompt).toContain("After");
+      expect(prompt).toContain("current state");
+      expect(prompt).toContain("target state");
+      expect(prompt).toContain("satisfies the user's request at a glance");
+    }
+  });
+
   it("triage planning prompt is sourced from workflow IR without an engine duplicate", () => {
     const corePrompt = resolveAgentPrompt("triage");
     const planningPrompt = resolvePlanningPromptFromIr(BUILTIN_CODING_WORKFLOW_IR);

packages/core/src/agent-prompts.ts

@@ -218,21 +218,24 @@ The tool prevents your session from being killed by the inactivity watchdog duri
 FNXC:FastPlanning 2026-06-28-17:05:
 Fast mode resolves the selected workflow's planning-fast seam before falling back to this built-in prompt.
 Keep the prompt lean, but preserve mandatory planning contracts: duplicate search, FN-5893 surface invariants, explicit workflow routing, forensic artifact paths, and no-commit decision specs.
+
+FNXC:FastPlanning 2026-07-04-16:25:
+Fast mode skips heavyweight planning ceremony, but every generated task still needs the same glanceable Before → After Transformation section as standard planning so operators can validate intent quickly.
 */
 const FAST_TRIAGE_PROMPT_TEXT = `You are a task specification agent for "fn". This task is running in **fast mode**.
 
-Write a lean, executable PROMPT.md quickly. Preserve safety-critical gates, but skip heavyweight ceremony, review scoring, and proactive subtask analysis.
+Write a lean, executable PROMPT.md quickly. Preserve safety gates, but skip heavyweight ceremony, review scoring, and proactive subtask analysis.
 
 ## Fast-mode priorities
-- Read only source/docs needed to make the spec precise; keep prose brief with concrete file paths, commands, and expected outcomes.
-- Do not expand scope. If work is already covered, report the duplicate instead of writing a new spec.
+- Read only source/docs needed for precision; keep prose brief with concrete file paths, commands, and outcomes.
+- Do not expand scope. If work is covered, report the duplicate instead of writing a new spec.
 - Preserve required safety sections for bugs, workflow routing, forensic tasks, and decision-only work.
 
 ## Duplicate check
 Before writing a spec, call \`fn_task_list\` for active work, then call \`fn_task_search\` with 2-4 targeted keyword phrases from the title/description, such as file paths, symptoms, and symbols. For any likely match in \`done\` or \`archived\`, call \`fn_task_show\` and inspect it before deciding. If an existing task covers the same work, do not write PROMPT.md; write exactly \`DUPLICATE: {existing-task-id}\`.
 
 ## Required PROMPT.md shape
-Write PROMPT.md with Mission, Dependencies, Context to Read First, File Scope, Steps, Documentation Requirements, Completion Criteria, Git Commit Convention, and Do NOT. In \`## Steps\`, every executable heading MUST use \`### Step N: <name>\` (for example, \`### Step 1: Preflight\`); Do not write bare \`### Preflight\` / \`### Implementation\` headings. Do not add review-level, triage subtask, or proactive subtask headings.
+Write PROMPT.md with Mission, Before → After Transformation, Dependencies, Context to Read First, File Scope, Steps, Documentation Requirements, Completion Criteria, Git Commit Convention, and Do NOT. Include \`## Before → After Transformation\` after Mission with concise Before and After bullets stating current state, target state, and why it satisfies the user's request at a glance. In \`## Steps\`, every executable heading MUST use \`### Step N: <name>\` (for example, \`### Step 1: Preflight\`); Do not write bare \`### Preflight\` / \`### Implementation\` headings. Do not add review-level, triage subtask, or proactive subtask headings.
 
 ## Surface Enumeration
 For bug fixes and UI-affordance add/remove tasks, the spec MUST include a \`## Surface Enumeration\` section. The workflow Plan Review gate validates this before execution when plan review is enabled.
@@ -303,6 +306,11 @@ Follow this structure exactly:
 
 {One paragraph: what you're building and why it matters}
 
+## Before → After Transformation
+
+- **Before:** {Briefly describe the current state, missing capability, broken behavior, or operator pain point}
+- **After:** {Briefly describe the target state and how it satisfies the user's request at a glance}
+
 ## Surface Enumeration
 
 {Required for bug-fix tasks and UI-affordance add/remove tasks (adding, removing, or restructuring icons, buttons, chevrons/arrows, toggles, badges, menu entries, click targets): a checklist enumerating every surface the fixed invariant must hold across. Include every provider/bridge for streaming and agent paths; desktop AND mobile breakpoints; empty/undefined/duplicate/populated data states; and every hook/component/module that shares the affected logic. For UI-affordance add/remove tasks, enumerate every component that renders the affordance by searching the codebase for the icon/class/testid — not just the component the user pointed at. Explicitly check for leftover shells after removal (empty buttons, orphaned click targets, now-unused wrappers, dangling aria-labels) across both desktop and mobile breakpoints. Use the canonical checklist in docs/testing.md as the starting point.}
@@ -427,6 +435,15 @@ If this task REMOVES existing functionality (deleting modules, settings, API end
 - This is mandatory for any net-negative change (more deletions than additions to existing files)
 \`\`\`
 
+## Transformation summary requirement
+
+Every normal implementation, documentation, or decision task definition MUST include \`## Before → After Transformation\` after \`## Mission\`. Keep it concise: use brief Before and After bullets (or equivalent short prose) that name the current state, the target state, and why that target satisfies the user's request at a glance.
+
+<!--
+FNXC:TriagePromptStructure 2026-07-04-16:20:
+Task definitions now carry a glanceable before-to-after transformation summary near the top so operators and reviewers can confirm the intended outcome before reading the full specification.
+-->
+
 ## Testing requirements
 
 The Testing & Verification step MUST require REAL automated tests — actual test

packages/engine/src/__tests__/triage.test.ts

@@ -22,6 +22,7 @@ const { mockReviewStep, mockCreateFnAgent } = vi.hoisted(() => ({
 }));
 
 const TRIAGE_POLICY_PROMPT = resolveAgentPrompt("triage");
+const STANDARD_PLANNING_PROMPT = builtinSeamPrompt("planning");
 const FAST_PLANNING_PROMPT = builtinSeamPrompt("planning-fast");
 const RENDERED_TRIAGE_POLICY_PROMPT = renderTriagePolicyPlaceholders(TRIAGE_POLICY_PROMPT, {});
 
@@ -733,6 +734,25 @@ describe("FN-5893 invariant regression wording", () => {
     "utf8",
   );
 
+  it("requires before-to-after transformation summaries in standard and fast planning prompts", () => {
+    for (const prompt of [
+      TRIAGE_POLICY_PROMPT,
+      STANDARD_PLANNING_PROMPT,
+      FAST_PLANNING_PROMPT,
+    ]) {
+      expect(prompt).toContain("## Before → After Transformation");
+      expect(prompt).toContain("Before");
+      expect(prompt).toContain("After");
+      expect(prompt).toContain("current state");
+      expect(prompt).toContain("target state");
+      expect(prompt).toContain("satisfies the user's request at a glance");
+    }
+
+    expect(STANDARD_PLANNING_PROMPT).toBe(TRIAGE_POLICY_PROMPT);
+    expect(FAST_PLANNING_PROMPT).not.toContain("## Review Level");
+    expect(FAST_PLANNING_PROMPT).not.toContain("## Proactive Subtask Breakdown");
+  });
+
   it("requires invariant-level regression coverage in standard, fast, and core triage prompts", () => {
     for (const prompt of [
       TRIAGE_POLICY_PROMPT,