Codex compatibility note:
- Invoke
$skill-name: load the registered .agents/skills/<name>/SKILL.md and execute it with Codex tools; no separate Skill tool or Claude mode switch is needed. When a workflow step or another skill names a skill, read that file in full yourself (every line, with no line limit; in parts only when output is cut off): a skill whose implicit invocation is off is never loaded for you. Author .claude/**; mirrors are generated.
- Track steps before execution and keep progress/evidence current. Explicit skill requests execute their protocol. Use the native question tool for required questions.
- Map foreign tool names to available host capabilities. A name mismatch needs no approval; an actually unavailable required capability stops with its step/evidence and a user question before adaptation.
- Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required
spawn_agent subagent(s) for that task.
- Preserve protocol order unless the user approves a deviation. Workflows use
$start-workflow: gates stay fixed; other steps may flex with logged reasons and evidence.
[BLOCKING] Execute skill steps in declared order. NEVER skip, reorder, or merge steps without explicit user approval. Inside a workflow run, $start-workflow → Workflow-step mode bounds this rule.
[BLOCKING] Before each step or sub-skill call, update todo tracking: set in_progress when step starts, set completed when step ends.
[BLOCKING] Every completed/skipped step MUST include brief evidence or explicit skip reason.
[BLOCKING] If Task tools are unavailable, create and maintain an equivalent step-by-step plan tracker with the same status transitions.
Quick Summary
Goal: Produce a defensible ranked ordering of 3+ planned work items using RICE, MoSCoW, or Value-Effort frameworks so the team works highest-value items first — every rank backed by a score and tech-agnostic rationale (value/effort/risk/impact).
Summary:
- 1. Track + detect/collect — track each declared step (
in_progress → completed + evidence); identify file/inline input and require ≥3 items; fewer → direct discussion, NEVER force a framework.
- 2. Select + score — honor a specified framework; otherwise RICE for quantitative data, MoSCoW for stakeholder alignment, Value-Effort 2x2 for a quick call; default RICE when unsure; apply exact criteria, formula, and scales.
- 3. Rank + report — rank by framework; emit a prioritized table with scores, tech-agnostic value/effort/risk/business-impact rationale, and Do-first/Plan-next/Defer recommendations.
- 4. Propagate + tie gate — when task files exist, record the integer
priority (1–999, lower first) + priority_label on EACH task, through $task-track when the task is tracked; near-tie/disagreement → ask user question tool for $llm-council vs accept, otherwise end without prompting.
Workflow:
- Collect Items — read files or parse inline list; require ≥3 items
- Select Framework — use RICE for quantitative data, MoSCoW for stakeholder alignment, or Value-Effort for a quick decision
- Score Each Item — apply exact framework criteria and calculate scores
- Rank & Report — emit prioritized table, rationale, and recommendations
- Propagate Priority — if task files exist, MANDATORY record integer
priority + priority_label on EACH task (tracked task: priority through $task-track)
- Tie Gate — near-tie →
ask user question tool ($llm-council vs accept); otherwise end without prompting
Key Rules:
- Require ≥3 items; fewer → discuss directly, NEVER force a framework.
- Honor a user-specified framework; otherwise default to RICE when unsure and ask when ambiguous.
AI-SDD Artifact Contract — M1–M7 are blocking: M1/M2 keep rationale tech-agnostic and source identifiers in evidence carriers; M3 uses logical IDs with abstract anchors; M4–M5 require unambiguous, rebuildable behavior; M6 makes review violations explicit; M7 requires demoable business outcomes.
MUST ATTENTION READ .claude/skills/shared/sdd-artifact-contract.md for full mandate and carrier rules.
- Tech-agnostic rationale (M1): justify every ranking by value, effort, risk, and business impact — NOT implementation technology. Rationale prose stays tech-agnostic per
spec-principles.md §3 in the reference-docs root (default docs/project-reference; a docsRoots.projectReference.path entry in docs/project-config.json overrides the path): no framework/product/language/design-pattern names; effort may cite effort points and relative complexity, never a named stack.
- Task propagation: when task files exist, record the rank as integer
priority (1–999, lower comes first) and the label as priority_label on EACH task; this is mandatory, not optional.
- Work record owner:
priority is a tracker-owned field (Records another skill authors). On a task with a tracking block, set it through $task-track --mode=maintain (update with priority) and edit only priority_label in the file; on an untracked task, write both keys directly. Ranking never changes status, readiness or assignee.
Apply critical/sequential thinking; every claim needs traced proof and confidence >80% to act.
Contents
Planned Work Prioritization
Use a data-driven framework → ranked list with scores + rationale.
When to Use
- Delivery planning needs ordered planned work (≥3 items)
- Stakeholder priority ranking needs justification
- Feature roadmap needs objective ordering
- Competing features or initiatives need comparison
When NOT to Use
- Fewer than 3 items → discuss directly
- Creating tasks or stories → use
work-item --mode=refine or work-item --mode=story
- Full product strategy → use
initiative then work-item --mode=refine
Phase 0: Detect & Prepare
- Input: ≥3 planned work items (tasks, features, user stories), from
tasks/ under the team-artifacts root (default team-artifacts; a docsRoots.teamArtifacts.path entry in docs/project-config.json overrides the path), a user path, or inline descriptions.
- Record any specified framework and available quantitative, stakeholder-alignment, or quick-decision signal.
- Fewer than 3 items → ask for more or discuss directly; NEVER force a framework.
Workflow
Collect items
- If a file path is provided, read items from files.
- If an inline list is provided, parse items from the user message.
- If fewer than 3 items, ask for more or suggest direct discussion.
Select framework using decision tree:
IF quantitative data available (reach, metrics) -> RICE
IF stakeholder alignment needed (must/should/could) -> MoSCoW
IF quick decision needed (2 axes only) -> Value-Effort 2x2
IF user specifies framework -> use that framework
IF unsure -> ask user, default RICE
Score each item using selected framework:
RICE:
Score = (Reach x Impact x Confidence) / Effort
Reach: Users affected per quarter (number)
Impact: 0.25 (minimal) | 0.5 (low) | 1 (medium) | 2 (high) | 3 (massive)
Confidence: 0.5 (low) | 0.8 (medium) | 1.0 (high)
Effort: Effort points (1, 2, 3, 5, 8, 13, 21)
MoSCoW:
Must Have: Critical for release, non-negotiable
Should Have: Important but not vital, workarounds exist
Could Have: Desirable, include if capacity allows
Won't Have: Out of scope for this cycle
Value-Effort 2x2:
High Value + Low Effort = Quick Wins (do first)
High Value + High Effort = Strategic (plan carefully)
Low Value + Low Effort = Fill-ins (if time permits)
Low Value + High Effort = Time Sinks (avoid)
Rank items by score (descending for RICE, category for MoSCoW, quadrant for V-E)
Output prioritized list with scores and rationale
IF task files exist -> MANDATORY priority propagation: record the resulting priority on EACH task — the rank as integer priority (1-999, lower comes first) and the label as priority_label (e.g. Must Have / Should Have / Could Have / Won't Have, or the framework's category). An untracked task takes both keys directly in its frontmatter; a task with a tracking block takes priority through $task-track --mode=maintain and only priority_label by file edit. When an existing task holds a label in priority and a number in rank, move the label to priority_label and the number to priority, and remove rank. Never leave this optional when task files exist: a task without its priority is incomplete, and downstream consumers (work-item --mode=mockup header badge, feature-presentation Scope & planned work slide) read priority FROM the task frontmatter. Update every ranked task, not just the standalone planned-work file.
Output Format
## Prioritized Planned Work
**Framework:** [RICE | MoSCoW | Value-Effort]
**Date:** [YYMMDD]
**Items scored:** [count]
### Rankings
| Rank | Item | Score | Rationale |
| ---- | --------- | ----- | --------------------------------------------------- |
| 1 | Feature A | 45.0 | High reach (5000), high impact (3), high confidence |
| 2 | Feature B | 12.0 | Medium reach (2000), medium impact, low effort |
| 3 | Feature C | 2.5 | Low reach, minimal impact, high effort |
### Recommendations
- **Do first:** [top items]
- **Plan next:** [medium items]
- **Defer:** [low items with reasoning]
Examples
Example 1: RICE scoring of 5 features
Input: "Prioritize: SSO login, dark mode, export to PDF, email notifications, bulk import"
Output:
| Rank |
Feature |
Reach |
Impact |
Conf |
Effort |
RICE |
| 1 |
Email notifications |
5000 |
2 |
0.8 |
1 |
8000 |
| 2 |
SSO login |
2000 |
3 |
0.8 |
3 |
1600 |
| 3 |
Bulk import |
500 |
2 |
1.0 |
1 |
1000 |
| 4 |
Export to PDF |
1000 |
1 |
0.8 |
2 |
400 |
| 5 |
Dark mode |
3000 |
0.5 |
0.5 |
2 |
375 |
Example 2: MoSCoW categorization
Input: "Categorize for Q1 release: payment gateway, admin dashboard redesign, API rate limiting, user avatars, audit logs"
Output:
- Must Have: Payment gateway (revenue-critical), API rate limiting (security)
- Should Have: Audit logs (compliance, workaround exists with manual exports)
- Could Have: Admin dashboard redesign (improves efficiency but current works)
- Won't Have: User avatars (nice-to-have, defer to Q2)
Optional Escalation: $llm-council on Ties
Gate evaluation: After producing prioritized planned work (per ## Workflow step output), inspect ranking output:
- Top-2 RICE scores within 15% of each other → gate fires
- Explicit MoSCoW tie (≥2 items in same Must/Should/Could band with material scope overlap) → gate fires
- Multi-stakeholder disagreement flagged in input → gate fires
- None of the above → gate does NOT fire; skill ends without prompting
MANDATORY ATTENTION — when the gate fires, you MUST use ask user question tool to present these options (identical preamble pattern to architecture --mode=design's ## Next Steps MANDATORY ATTENTION block):
- "Escalate to $llm-council (Recommended)" — Tie/disagreement detected. Run 11 sub-agent council (5 advisors + 5 reviewers + chairman). Council's Contrarian + Outsider lenses are well-suited to multi-task ranking ties. Cheaper alternatives:
$why-review, $plan --mode=validate (use these instead if the tie is narrow but stakes are routine).
- "Skip — accept current ranking" — Acknowledge the tie; proceed with current ranking.
If gate does NOT fire, the prioritization decision stands; do NOT prompt.
Related Skills
| Skill |
When to use instead |
initiative |
Capturing product intent upfront |
work-item --mode=story |
Breaking tasks into user stories |
work-item --mode=refine |
Refining initiatives into tasks |
Prompt-Enhance Closing Anchors
- IMPORTANT MUST ATTENTION follow declared step order for this skill; NEVER skip, reorder, or merge steps without explicit user approval; inside a workflow run,
$start-workflow → Workflow-step mode bounds this
- IMPORTANT MUST ATTENTION for every step/sub-skill call: set
in_progress before execution, set completed after execution
- IMPORTANT MUST ATTENTION every skipped step MUST include explicit reason; every completed step MUST include concise evidence
- IMPORTANT MUST ATTENTION if Task tools unavailable, maintain an equivalent step-by-step plan tracker with synchronized statuses
Closing Reminders
IMPORTANT MUST ATTENTION Goal: Produce a defensible ranked ordering of 3+ planned work items using RICE, MoSCoW, or Value-Effort frameworks so the team works highest-value items first — every rank backed by a score and tech-agnostic rationale (value/effort/risk/impact).
Protocols in force (concise digest of the SYNC/shared blocks this skill carries):
- IMPORTANT MUST ATTENTION require 3+ items BEFORE ranking; fewer than 3 → discuss directly, NEVER force a framework — why: ranking 1-2 items adds ceremony without signal
- IMPORTANT MUST ATTENTION select the framework by the decision tree — RICE for quantitative data, MoSCoW for stakeholder must/should/could alignment, Value-Effort 2x2 for a quick call; default RICE when unsure, ask the user when ambiguous — why: matching framework to the decision type is what makes the ranking defensible
- IMPORTANT MUST ATTENTION emit the prioritized table (scores + rationale) AND Do-first/Plan-next/Defer recommendations — NEVER stop at raw scores — why: the consumable output is the ranked table plus an action call, not a number column
- IMPORTANT MUST ATTENTION keep every rationale tech-agnostic per M1 — justify by value/effort/risk/business impact, NEVER by named stack/framework/product/language/design-pattern; effort may cite effort points + relative complexity only — why: spec-principles §3 BLOCKING, a tech-named rationale leaks implementation into a priority call
- IMPORTANT MUST ATTENTION score with the EXACT framework formula (RICE = Reach×Impact×Confidence ÷ Effort, fixed Impact/Confidence scales, Effort in effort points), then rank descending (RICE) / by band (MoSCoW) / by quadrant (V-E) — NEVER invent ad-hoc scores — why: a defensible rank needs a reproducible number
- IMPORTANT MUST ATTENTION on a near-tie (top-2 RICE within 15%, same-band MoSCoW overlap, flagged stakeholder disagreement) the gate FIRES — use
ask user question tool to offer $llm-council escalation vs. accepting the ranking; if the gate does NOT fire, end WITHOUT prompting — why: tie-breaking is a judgment call the user owns, but a clear winner needs no interruption
- IMPORTANT MUST ATTENTION break work into small todo tasks using todo tracking BEFORE starting; mark one
in_progress, completed immediately after evidence
- IMPORTANT MUST ATTENTION search codebase/artifacts for 3+ similar patterns before creating new structure; evaluate pattern FIT (same constraints/scope) before copying a nearby example — why: closest example ≠ matching preconditions
- IMPORTANT MUST ATTENTION cite
file:line evidence for every claim (confidence >80% to act, <60% DO NOT recommend); NEVER present a guess as fact
- IMPORTANT MUST ATTENTION when task files exist, propagating the ranking onto EACH task (integer
priority 1-999 + priority_label; on a tracked task priority goes through $task-track) is MANDATORY, not optional — do it after ranking; grep downstream consumers before changing any priority field — why: downstream consumers (work-item --mode=mockup header, feature-presentation Scope & planned work slide) read priority from task frontmatter, and stale/absent priority refs cascade silently
- IMPORTANT MUST ATTENTION add a final review todo task to verify work quality
Anti-Rationalization:
| Evasion |
Rebuttal |
| "Only 2 items, just rank them" |
Below the 3-item floor → discuss directly; a framework adds ceremony, not signal |
| "I'll cite the framework in the rationale" |
Tech-agnostic per M1 — justify by value/effort/risk only, never by named stack |
| "Scores are close enough, I'll pick" |
Near-tie fires the gate → ask user question tool for $llm-council, never silently break |
| "RICE feels right, skip the formula" |
Apply the EXACT formula with fixed scales — a defensible rank needs a number |
| "The planned-work file has the ranking, tasks don't need it" |
When task files exist, priority write-back to each task frontmatter is MANDATORY — mockup + presentation read priority from the task, not the planned work |
| "Already know the patterns" |
Show file:line evidence — no proof = no search |
| [TASK-PLANNING] Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using todo tracking. |
|
[IMPORTANT] Analyze how big the task is and break it into many small todo tasks systematically before starting — this is very important.
[IMPORTANT] Use todo tracking to break ALL work into small tasks BEFORE starting. This prevents context loss from long files. Keep task depth proportional to the work.
1---2name: prioritize-43description: '[Project Management] Use when a workflow step or the user asks for planned-work prioritization using RICE, MoSCoW or Value-Effort ranking.'4metadata:5 source: https://skillmd.com/skills/duc01226/prioritize-46---78> Codex compatibility note:9> - Invoke `$skill-name`: load the registered `.agents/skills/<name>/SKILL.md` and execute it with Codex tools; no separate `Skill` tool or Claude mode switch is needed. When a workflow step or another skill names a skill, read that file in full yourself (every line, with no line limit; in parts only when output is cut off): a skill whose implicit invocation is off is never loaded for you. Author `.claude/**`; mirrors are generated.10> - Track steps before execution and keep progress/evidence current. Explicit skill requests execute their protocol. Use the native question tool for required questions.11> - Map foreign tool names to available host capabilities. A name mismatch needs no approval; an actually unavailable required capability stops with its step/evidence and a user question before adaptation.12> - Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required `spawn_agent` subagent(s) for that task.13> - Preserve protocol order unless the user approves a deviation. Workflows use `$start-workflow`: gates stay fixed; other steps may flex with logged reasons and evidence.14<!-- PROMPT-ENHANCE:STEP-TASK-ANCHOR:START -->1516> **[BLOCKING]** Execute skill steps in declared order. NEVER skip, reorder, or merge steps without explicit user approval. Inside a workflow run, `$start-workflow` → Workflow-step mode bounds this rule.17> **[BLOCKING]** Before each step or sub-skill call, update todo tracking: set `in_progress` when step starts, set `completed` when step ends.18> **[BLOCKING]** Every completed/skipped step MUST include brief evidence or explicit skip reason.19> **[BLOCKING]** If Task tools are unavailable, create and maintain an equivalent step-by-step plan tracker with the same status transitions.2021<!-- PROMPT-ENHANCE:STEP-TASK-ANCHOR:END -->2223## Quick Summary2425**Goal:** Produce a defensible ranked ordering of 3+ planned work items using RICE, MoSCoW, or Value-Effort frameworks so the team works highest-value items first — every rank backed by a score and tech-agnostic rationale (value/effort/risk/impact).2627**Summary:**2829- **1. Track + detect/collect** — track each declared step (`in_progress` → `completed` + evidence); identify file/inline input and require ≥3 items; fewer → direct discussion, NEVER force a framework.30- **2. Select + score** — honor a specified framework; otherwise RICE for quantitative data, MoSCoW for stakeholder alignment, Value-Effort 2x2 for a quick call; default RICE when unsure; apply exact criteria, formula, and scales.31- **3. Rank + report** — rank by framework; emit a prioritized table with scores, tech-agnostic value/effort/risk/business-impact rationale, and Do-first/Plan-next/Defer recommendations.32- **4. Propagate + tie gate** — when task files exist, record the integer `priority` (1–999, lower first) + `priority_label` on EACH task, through `$task-track` when the task is tracked; near-tie/disagreement → `ask user question tool` for `$llm-council` vs accept, otherwise end without prompting.3334**Workflow:**35361. **Collect Items** — read files or parse inline list; require ≥3 items372. **Select Framework** — use RICE for quantitative data, MoSCoW for stakeholder alignment, or Value-Effort for a quick decision383. **Score Each Item** — apply exact framework criteria and calculate scores394. **Rank & Report** — emit prioritized table, rationale, and recommendations405. **Propagate Priority** — if task files exist, MANDATORY record integer `priority` + `priority_label` on EACH task (tracked task: `priority` through `$task-track`)416. **Tie Gate** — near-tie → `ask user question tool` (`$llm-council` vs accept); otherwise end without prompting4243**Key Rules:**4445- Require ≥3 items; fewer → discuss directly, NEVER force a framework.46- Honor a user-specified framework; otherwise default to RICE when unsure and ask when ambiguous.4748> **AI-SDD Artifact Contract** — M1–M7 are blocking: M1/M2 keep rationale tech-agnostic and source identifiers in evidence carriers; M3 uses logical IDs with abstract anchors; M4–M5 require unambiguous, rebuildable behavior; M6 makes review violations explicit; M7 requires demoable business outcomes.49> MUST ATTENTION READ `.claude/skills/shared/sdd-artifact-contract.md` for full mandate and carrier rules.5051- **Tech-agnostic rationale (M1):** justify every ranking by value, effort, risk, and business impact — NOT implementation technology. Rationale prose stays tech-agnostic per `spec-principles.md` §3 in the reference-docs root (default `docs/project-reference`; a `docsRoots.projectReference.path` entry in `docs/project-config.json` overrides the path): no framework/product/language/design-pattern names; effort may cite effort points and relative complexity, never a named stack.52- **Task propagation:** when task files exist, record the rank as integer `priority` (1–999, lower comes first) and the label as `priority_label` on EACH task; this is mandatory, not optional.53- **Work record owner:** `priority` is a tracker-owned field ([Records another skill authors](../task-track/references/integration-guide.md#records-another-skill-authors)). On a task with a `tracking` block, set it through `$task-track --mode=maintain` (`update` with `priority`) and edit only `priority_label` in the file; on an untracked task, write both keys directly. Ranking never changes `status`, readiness or assignee.5455Apply critical/sequential thinking; every claim needs traced proof and confidence >80% to act.5657## Contents5859- [Quick Summary](#quick-summary)60- [Planned Work Prioritization](#planned-work-prioritization)61- [When to Use](#when-to-use)62- [When NOT to Use](#when-not-to-use)63- [Phase 0: Detect & Prepare](#phase-0-detect--prepare)64- [Workflow](#workflow)65- [Output Format](#output-format)66- [Examples](#examples)67- [Optional Escalation: $llm-council on Ties](#optional-escalation-llm-council-on-ties)68- [Related Skills](#related-skills)69- [Prompt-Enhance Closing Anchors](#prompt-enhance-closing-anchors)70- [Closing Reminders](#closing-reminders)7172# Planned Work Prioritization7374Use a data-driven framework → ranked list with scores + rationale.7576## When to Use7778- Delivery planning needs ordered planned work (≥3 items)79- Stakeholder priority ranking needs justification80- Feature roadmap needs objective ordering81- Competing features or initiatives need comparison8283## When NOT to Use8485- Fewer than 3 items → discuss directly86- Creating tasks or stories → use `work-item --mode=refine` or `work-item --mode=story`87- Full product strategy → use `initiative` then `work-item --mode=refine`8889## Phase 0: Detect & Prepare9091- Input: ≥3 planned work items (tasks, features, user stories), from `tasks/` under the team-artifacts root (default `team-artifacts`; a `docsRoots.teamArtifacts.path` entry in `docs/project-config.json` overrides the path), a user path, or inline descriptions.92- Record any specified framework and available quantitative, stakeholder-alignment, or quick-decision signal.93- Fewer than 3 items → ask for more or discuss directly; NEVER force a framework.9495## Workflow96971. **Collect items**98 - If a file path is provided, read items from files.99 - If an inline list is provided, parse items from the user message.100 - If fewer than 3 items, ask for more or suggest direct discussion.1011022. **Select framework** using decision tree:103104 ```105 IF quantitative data available (reach, metrics) -> RICE106 IF stakeholder alignment needed (must/should/could) -> MoSCoW107 IF quick decision needed (2 axes only) -> Value-Effort 2x2108 IF user specifies framework -> use that framework109 IF unsure -> ask user, default RICE110 ```1111123. **Score each item** using selected framework:113114 **RICE:**115116 ```117 Score = (Reach x Impact x Confidence) / Effort118119 Reach: Users affected per quarter (number)120 Impact: 0.25 (minimal) | 0.5 (low) | 1 (medium) | 2 (high) | 3 (massive)121 Confidence: 0.5 (low) | 0.8 (medium) | 1.0 (high)122 Effort: Effort points (1, 2, 3, 5, 8, 13, 21)123 ```124125 **MoSCoW:**126127 ```128 Must Have: Critical for release, non-negotiable129 Should Have: Important but not vital, workarounds exist130 Could Have: Desirable, include if capacity allows131 Won't Have: Out of scope for this cycle132 ```133134 **Value-Effort 2x2:**135136 ```137 High Value + Low Effort = Quick Wins (do first)138 High Value + High Effort = Strategic (plan carefully)139 Low Value + Low Effort = Fill-ins (if time permits)140 Low Value + High Effort = Time Sinks (avoid)141 ```1421434. **Rank items** by score (descending for RICE, category for MoSCoW, quadrant for V-E)1441455. **Output** prioritized list with scores and rationale1461476. **IF task files exist** -> **MANDATORY priority propagation**: record the resulting priority on EACH task — the rank as integer `priority` (1-999, lower comes first) and the label as `priority_label` (e.g. Must Have / Should Have / Could Have / Won't Have, or the framework's category). An untracked task takes both keys directly in its frontmatter; a task with a `tracking` block takes `priority` through `$task-track --mode=maintain` and only `priority_label` by file edit. When an existing task holds a label in `priority` and a number in `rank`, move the label to `priority_label` and the number to `priority`, and remove `rank`. Never leave this optional when task files exist: a task without its priority is incomplete, and downstream consumers (`work-item --mode=mockup` header badge, `feature-presentation` Scope & planned work slide) read priority FROM the task frontmatter. Update every ranked task, not just the standalone planned-work file.148149## Output Format150151```markdown152## Prioritized Planned Work153154**Framework:** [RICE | MoSCoW | Value-Effort]155**Date:** [YYMMDD]156**Items scored:** [count]157158### Rankings159160| Rank | Item | Score | Rationale |161| ---- | --------- | ----- | --------------------------------------------------- |162| 1 | Feature A | 45.0 | High reach (5000), high impact (3), high confidence |163| 2 | Feature B | 12.0 | Medium reach (2000), medium impact, low effort |164| 3 | Feature C | 2.5 | Low reach, minimal impact, high effort |165166### Recommendations167168- **Do first:** [top items]169- **Plan next:** [medium items]170- **Defer:** [low items with reasoning]171```172173## Examples174175### Example 1: RICE scoring of 5 features176177**Input:** "Prioritize: SSO login, dark mode, export to PDF, email notifications, bulk import"178179**Output:**180181| Rank | Feature | Reach | Impact | Conf | Effort | RICE |182| ---- | ------------------- | ----- | ------ | ---- | ------ | ---- |183| 1 | Email notifications | 5000 | 2 | 0.8 | 1 | 8000 |184| 2 | SSO login | 2000 | 3 | 0.8 | 3 | 1600 |185| 3 | Bulk import | 500 | 2 | 1.0 | 1 | 1000 |186| 4 | Export to PDF | 1000 | 1 | 0.8 | 2 | 400 |187| 5 | Dark mode | 3000 | 0.5 | 0.5 | 2 | 375 |188189### Example 2: MoSCoW categorization190191**Input:** "Categorize for Q1 release: payment gateway, admin dashboard redesign, API rate limiting, user avatars, audit logs"192193**Output:**194195- **Must Have:** Payment gateway (revenue-critical), API rate limiting (security)196- **Should Have:** Audit logs (compliance, workaround exists with manual exports)197- **Could Have:** Admin dashboard redesign (improves efficiency but current works)198- **Won't Have:** User avatars (nice-to-have, defer to Q2)199200## Optional Escalation: $llm-council on Ties201202**Gate evaluation:** After producing prioritized planned work (per `## Workflow` step output), inspect ranking output:203204- Top-2 RICE scores within 15% of each other → gate fires205- Explicit MoSCoW tie (≥2 items in same Must/Should/Could band with material scope overlap) → gate fires206- Multi-stakeholder disagreement flagged in input → gate fires207- None of the above → gate does NOT fire; skill ends without prompting208209**MANDATORY ATTENTION** — when the gate fires, you MUST use `ask user question tool` to present these options (identical preamble pattern to `architecture --mode=design`'s `## Next Steps` MANDATORY ATTENTION block):210211- **"Escalate to $llm-council (Recommended)"** — Tie/disagreement detected. Run 11 sub-agent council (5 advisors + 5 reviewers + chairman). Council's Contrarian + Outsider lenses are well-suited to multi-task ranking ties. Cheaper alternatives: `$why-review`, `$plan --mode=validate` (use these instead if the tie is narrow but stakes are routine).212- **"Skip — accept current ranking"** — Acknowledge the tie; proceed with current ranking.213214If gate does NOT fire, the prioritization decision stands; do NOT prompt.215216## Related Skills217218| Skill | When to use instead |219| ----------------- | ---------------------------------- |220| `initiative` | Capturing product intent upfront |221| `work-item --mode=story` | Breaking tasks into user stories |222| `work-item --mode=refine` | Refining initiatives into tasks |223224---225226<!-- PROMPT-ENHANCE:STEP-TASK-CLOSING:START -->227228## Prompt-Enhance Closing Anchors229230- **IMPORTANT MUST ATTENTION** follow declared step order for this skill; NEVER skip, reorder, or merge steps without explicit user approval; inside a workflow run, `$start-workflow` → Workflow-step mode bounds this231- **IMPORTANT MUST ATTENTION** for every step/sub-skill call: set `in_progress` before execution, set `completed` after execution232- **IMPORTANT MUST ATTENTION** every skipped step MUST include explicit reason; every completed step MUST include concise evidence233- **IMPORTANT MUST ATTENTION** if Task tools unavailable, maintain an equivalent step-by-step plan tracker with synchronized statuses234235<!-- PROMPT-ENHANCE:STEP-TASK-CLOSING:END -->236237## Closing Reminders238239**IMPORTANT MUST ATTENTION Goal:** Produce a defensible ranked ordering of 3+ planned work items using RICE, MoSCoW, or Value-Effort frameworks so the team works highest-value items first — every rank backed by a score and tech-agnostic rationale (value/effort/risk/impact).240241**Protocols in force (concise digest of the SYNC/shared blocks this skill carries):**242243244- **IMPORTANT MUST ATTENTION** require 3+ items BEFORE ranking; fewer than 3 → discuss directly, NEVER force a framework — why: ranking 1-2 items adds ceremony without signal245- **IMPORTANT MUST ATTENTION** select the framework by the decision tree — RICE for quantitative data, MoSCoW for stakeholder must/should/could alignment, Value-Effort 2x2 for a quick call; default RICE when unsure, ask the user when ambiguous — why: matching framework to the decision type is what makes the ranking defensible246- **IMPORTANT MUST ATTENTION** emit the prioritized table (scores + rationale) AND Do-first/Plan-next/Defer recommendations — NEVER stop at raw scores — why: the consumable output is the ranked table plus an action call, not a number column247- **IMPORTANT MUST ATTENTION** keep every rationale tech-agnostic per M1 — justify by value/effort/risk/business impact, NEVER by named stack/framework/product/language/design-pattern; effort may cite effort points + relative complexity only — why: spec-principles §3 BLOCKING, a tech-named rationale leaks implementation into a priority call248- **IMPORTANT MUST ATTENTION** score with the EXACT framework formula (RICE = Reach×Impact×Confidence ÷ Effort, fixed Impact/Confidence scales, Effort in effort points), then rank descending (RICE) / by band (MoSCoW) / by quadrant (V-E) — NEVER invent ad-hoc scores — why: a defensible rank needs a reproducible number249- **IMPORTANT MUST ATTENTION** on a near-tie (top-2 RICE within 15%, same-band MoSCoW overlap, flagged stakeholder disagreement) the gate FIRES — use `ask user question tool` to offer `$llm-council` escalation vs. accepting the ranking; if the gate does NOT fire, end WITHOUT prompting — why: tie-breaking is a judgment call the user owns, but a clear winner needs no interruption250- **IMPORTANT MUST ATTENTION** break work into small todo tasks using todo tracking BEFORE starting; mark one `in_progress`, `completed` immediately after evidence251- **IMPORTANT MUST ATTENTION** search codebase/artifacts for 3+ similar patterns before creating new structure; evaluate pattern FIT (same constraints/scope) before copying a nearby example — why: closest example ≠ matching preconditions252- **IMPORTANT MUST ATTENTION** cite `file:line` evidence for every claim (confidence >80% to act, <60% DO NOT recommend); NEVER present a guess as fact253- **IMPORTANT MUST ATTENTION** when task files exist, propagating the ranking onto EACH task (integer `priority` 1-999 + `priority_label`; on a tracked task `priority` goes through `$task-track`) is MANDATORY, not optional — do it after ranking; grep downstream consumers before changing any priority field — why: downstream consumers (`work-item --mode=mockup` header, `feature-presentation` Scope & planned work slide) read priority from task frontmatter, and stale/absent priority refs cascade silently254- **IMPORTANT MUST ATTENTION** add a final review todo task to verify work quality255256**Anti-Rationalization:**257258| Evasion | Rebuttal |259| ------------------------------------ | ----------------------------------------------------------------------------- |260| "Only 2 items, just rank them" | Below the 3-item floor → discuss directly; a framework adds ceremony, not signal |261| "I'll cite the framework in the rationale" | Tech-agnostic per M1 — justify by value/effort/risk only, never by named stack |262| "Scores are close enough, I'll pick" | Near-tie fires the gate → `ask user question tool` for `$llm-council`, never silently break |263| "RICE feels right, skip the formula" | Apply the EXACT formula with fixed scales — a defensible rank needs a number |264| "The planned-work file has the ranking, tasks don't need it" | When task files exist, priority write-back to each task frontmatter is MANDATORY — mockup + presentation read priority from the task, not the planned work |265| "Already know the patterns" | Show `file:line` evidence — no proof = no search |266**[TASK-PLANNING]** Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using todo tracking.267268> **[IMPORTANT]** Analyze how big the task is and break it into many small todo tasks systematically before starting — this is very important.269270> **[IMPORTANT]** Use todo tracking to break ALL work into small tasks BEFORE starting. This prevents context loss from long files. Keep task depth proportional to the work.