Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Create, track, and archive projects linked to goals. The bridge between goals and daily tasks. Use for project creation, status dashboards, and archiving completed work.
.claude/skills/ballred-project/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 45% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 165% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 173% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 26% | 0% |
Create, track, and archive projects that bridge the gap between goals and daily tasks.
/project # Interactive: create new project or view status
/project new # Create a new project
/project status # Dashboard of all active projects
/project archive <name> # Archive a completed project/project or /project newCreates a new project folder with a CLAUDE.md context file, interactively linked to a goal.
Steps:
Goals/1. Yearly Goals.md to list available goalsProjects/<ProjectName>/CLAUDE.md with structure below[[Projects/<ProjectName>]] reference in the yearly goals fileProject CLAUDE.md Template:
markdown# Project: <Name> ## Overview [Brief description of what this project achieves] ## Goal Link Supports: [[1. Yearly Goals#<Goal Name>]] ## Status - **Phase:** Planning | Active | Review | Complete - **Progress:** 0% - **Started:** <date> - **Target:** <date> ## Key Decisions - [Decision 1] - [Date] - [Rationale] ## Next Actions - [ ] [First concrete step] - [ ] [Second step] ## Notes [Running log of updates, blockers, learnings]
/project statusScans all Projects/*/CLAUDE.md files and displays a dashboard.
Steps:
Projects/*/CLAUDE.mdOutput Format:
markdown## Project Dashboard | Project | Phase | Progress | Goal | Next Action | |---------|-------|----------|------|-------------| | ProjectA | Active | 60% | [[Goal 1]] | Review PR | | ProjectB | Planning | 10% | [[Goal 3]] | Draft spec | ### Summary - Active projects: N - Total progress (weighted): X% - Projects without goal link: [list] - Stalled projects (no update in 14+ days): [list]
/project archive <name>Moves a completed project to the archives.
Steps:
Projects/<name>/ existsmv Projects/<name> Archives/Projects/<name>Archives/Projects/ directory if it doesn't existProjects/LearnSpanish/Projects are the critical middle layer:
Goals/1. Yearly Goals.md <- "What I want to achieve"
|
v
Projects/*/CLAUDE.md <- "How I'll achieve it" (THIS SKILL)
|
v
Daily Notes/*.md <- "What I'm doing today"When creating tasks in daily notes, reference the project:
markdown- [ ] Draft API spec — [[Projects/MyApp/CLAUDE.md|MyApp]]
TaskCreate:
subject: "Read yearly goals"
description: "Load goals for project linking"
activeForm: "Reading yearly goals..."
TaskCreate:
subject: "Create project structure"
description: "Create folder and CLAUDE.md for new project"
activeForm: "Creating project structure..."
TaskCreate:
subject: "Link project to goal"
description: "Add project reference to yearly goals file"
activeForm: "Linking project to goal..."TaskCreate:
subject: "Scan project files"
description: "Glob and read all Projects/*/CLAUDE.md files"
activeForm: "Scanning project files..."
TaskCreate:
subject: "Generate dashboard"
description: "Compile status dashboard from project data"
activeForm: "Generating project dashboard..."Mark each task in_progress when starting, completed when done.
Works with:
/daily - Surface project next-actions in morning routine/weekly - Project status in weekly review/goal-tracking - Project progress feeds goal calculations/onboard - Discover and load project context/push - Commit project changes| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 11,434 | 16,082 | +41% | 1 | 1 | 0% | 2,270 | 4,373 | +93% | 0 | 0 | — |
case-02 | fail→fail | 5,261 | 5,670 | +8% | 1 | 1 | 0% | 931 | 1,525 | +64% | 0 | 0 | — |
case-03 | fail→fail | 7,801 | 5,477 | -30% | 1 | 1 | 0% | 1,385 | 1,375 | -1% | 0 | 0 | — |
case-04 | fail→fail | 9,696 | 6,924 | -29% | 1 | 1 | 0% | 2,002 | 1,568 | -22% | 0 | 0 | — |
case-05 | fail→fail | 7,533 | 7,914 | +5% | 1 | 1 | 0% | 1,389 | 1,752 | +26% | 0 | 0 | — |
case-06 | fail→fail | 3,504 | 14,342 | +309% | 1 | 1 | 0% | 537 | 1,549 | +188% | 0 | 0 | — |
case-07 | pass→pass | 10,333 | 3,818 | -63% | 1 | 1 | 0% | 1,606 | 1,723 | +7% | 0 | 0 | — |
case-08 | fail→pass | 7,110 | 3,085 | -57% | 1 | 1 | 0% | 1,154 | 1,673 | +45% | 0 | 0 | — |
case-14 | fail→fail | 11,840 | 7,870 | -34% | 1 | 1 | 0% | 2,039 | 1,572 | -23% | 0 | 0 | — |
case-09 | fail→fail | 10,732 | 12,317 | +15% | 1 | 1 | 0% | 2,163 | 2,619 | +21% | 0 | 0 | — |
case-10 | fail→fail | 4,935 | 4,466 | -10% | 1 | 1 | 0% | 903 | 1,372 | +52% | 0 | 0 | — |
case-11 | fail→pass | 8,554 | 2,669 | -69% | 1 | 1 | 0% | 1,515 | 1,635 | +8% | 0 | 0 | — |
case-12 | fail→pass | 6,046 | 12,621 | +109% | 1 | 1 | 0% | 959 | 2,540 | +165% | 0 | 0 | — |
case-13 | fail→pass | 3,031 | 1,207 | -60% | 1 | 1 | 0% | 470 | 1,281 | +173% | 0 | 0 | — |
case-15 | pass→pass | 5,931 | 1,227 | -79% | 1 | 1 | 0% | 979 | 1,277 | +30% | 0 | 0 | — |
case-16 | pass→pass | 9,324 | 3,808 | -59% | 1 | 1 | 0% | 1,660 | 1,726 | +4% | 0 | 0 | — |
case-17 | fail→pass | 6,854 | 2,559 | -63% | 1 | 1 | 0% | 1,251 | 1,573 | +26% | 0 | 0 | — |
case-18 | pass→pass | 3,725 | 1,499 | -60% | 1 | 1 | 0% | 615 | 1,393 | +127% | 0 | 0 | — |
case-19 | pass→pass | 11,613 | 3,805 | -67% | 1 | 1 | 0% | 1,867 | 1,836 | -2% | 0 | 0 | — |
case-20 | pass→pass | 4,191 | 5,963 | +42% | 1 | 1 | 0% | 631 | 2,149 | +241% | 0 | 0 | — |
case-21 | fail→fail | 18,131 | 4,523 | -75% | 1 | 1 | 0% | 1,207 | 1,338 | +11% | 0 | 0 | — |
case-22 | pass→pass | 2,930 | 8,036 | +174% | 1 | 1 | 0% | 363 | 2,589 | +613% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 12 counted toward the lift figure. The other 10 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +23 percentage points is the difference between those two pass rates over the 12 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.