Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Scores feature worthiness and enforces branch-size limits against overengineering. Use when evaluating whether a feature belongs in the current scope or branch.
.claude/skills/athola-scope-guard/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 65% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 142% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 314% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 103% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 175% | 0% |
> The cheapest feature is the one you never build. > Scope-guard weighs each proposed feature against its value, > its cost, and the branch budget, so only what is earned ships.
scope-guard:worthiness-scored))scope-guard:backlog-compared))scope-guard:budget-checked))scope-guard:decision-documented))Prevents overengineering by both Claude and human during the brainstorm, plan, and execute workflow. Forces explicit evaluation of every proposed feature against business value, opportunity cost, and branch constraints.
Core Belief: Not all features deserve implementation. Most ideas should be deferred to backlog until proven necessary.
Three Pillars:
Use the Worthiness formula:
(Business Value + Time Criticality + Risk Reduction) / (Complexity + Token Cost + Scope Drift)Verification: Run the command with --help flag to verify availability.
See decision-framework.md for details.
Thresholds:
Compare against docs/backlog/queue.md:
Default: 3 major features per branch
If at capacity, must drop existing feature, split to new branch, or justify override.
Watch for Yellow/Red zones:
See branch-management.md for monitoring.
scope-guard:worthiness-scored)Score each factor (1, 2, 3, 5, 8, 13):
Details: decision-framework.md
scope-guard:backlog-compared)docs/backlog/queue.md for existing itemsscope-guard:budget-checked)Count current features in branch. If at budget (default: 3), new feature requires:
scope-guard:decision-documented)Record outcome:
github-integration.md Steps 1-3
scope-guard:github-issue-created completegithub-integration.md Step 4. Prompt: "Creating a Discussion with full reasoning context. Y/n]" Publishing is the default. If the user explicitly declines, skip Discussion creation. If publishing fails, continue.
docs/backlog/queue.md with issue linkIMPORTANT: Deferral is NOT complete until a GitHub issue exists. This prevents context loss when branches are merged or abandoned.
Key Principles:
See anti-overengineering.md for full rules and red flags.
**Verification:** Run the command with `--help` flag to verify availability.
docs/backlog/
├── queue.md # Active ranked queue
└── archive/
├── ideas.md # Deferred feature ideas
├── optimizations.md # Deferred performance work
├── refactors.md # Deferred cleanup
└── abstractions.md # Deferred patternsVerification: Run the command with --help flag to verify availability.
When deferring, add to docs/backlog/queue.md:
markdown| Rank | Item | Worthiness | Added | Branch/Epic | Category | |------|------|------------|-------|-------------|----------| | 1 | [New item description] | 1.8 | 2025-12-08 | current-branch | idea |
Verification: Run the command with --help flag to verify availability.
Re-rank by Worthiness Score after adding.
At end of brainstorming, before documenting design:
Self-invoke prompt: "Before documenting this design, let me evaluate the proposed features with scope-guard."
Before finalizing implementation plan:
Self-invoke prompt: "Before finalizing this plan, let me verify scope with scope-guard."
Periodically during execution:
Self-invoke prompt: "This branch has grown significantly. Let me check scope-guard thresholds."
When evaluating a feature, create these todos:
scope-guard:worthiness-scoredscope-guard:backlog-comparedscope-guard:budget-checkedscope-guard:github-issue-created (MANDATORY if deferring - blocks step 5)scope-guard:decision-documentedNote: Step 4 (github-issue-created) is REQUIRED when deferring items. You cannot mark decision-documented complete without first completing github-issue-created for deferrals.
superpowers:brainstorming - Ideation workflow this guardssuperpowers:writing-plans - Planning workflow this validatesimbue:review-core - Review methodology patternimbue:karpathy-principles - Compact four-principle synthesis covering scope, simplicity, surgical edits, and verifiable goalsdocs/quality-gates.md#skill-level-quality-gate-composition for the full gate-skill federation graphformula (Business Value + Time Criticality + Risk Reduction) / (Complexity + Token Cost + Scope Drift) and a disposition: implement (>2.0), discuss (1.0-2.0), or defer (<1.0)
scope-guard:decision-documented is marked complete; the issue URL is recorded in the todo
requires dropping an existing one, splitting to a new branch, or an explicit written override with justification
15/25/30, days 3/7/7+); Yellow zone triggers a warning, Red zone blocks without justification
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 21,331 | 15,408 | -28% | 1 | 1 | 0% | 3,265 | 5,381 | +65% | 0 | 0 | — |
case-02 | fail→fail | 28,228 | 21,388 | -24% | 1 | 1 | 0% | 4,578 | 6,428 | +40% | 0 | 0 | — |
case-03 | fail→pass | 18,875 | 18,317 | -3% | 1 | 1 | 0% | 2,475 | 5,993 | +142% | 0 | 0 | — |
case-04 | pass→pass | 11,533 | 9,143 | -21% | 1 | 1 | 0% | 1,748 | 3,889 | +122% | 0 | 0 | — |
case-05 | fail→pass | 5,218 | 2,532 | -51% | 1 | 1 | 0% | 710 | 2,937 | +314% | 0 | 0 | — |
case-06 | pass→fail | 9,981 | 4,392 | -56% | 1 | 1 | 0% | 1,798 | 3,300 | +84% | 0 | 0 | — |
case-07 | fail→pass | 8,351 | 3,410 | -59% | 1 | 1 | 0% | 1,621 | 3,286 | +103% | 0 | 0 | — |
case-08 | fail→pass | 5,981 | 2,499 | -58% | 1 | 1 | 0% | 1,125 | 3,096 | +175% | 0 | 0 | — |
case-09 | fail→pass | 6,929 | 3,924 | -43% | 1 | 1 | 0% | 1,244 | 3,338 | +168% | 0 | 0 | — |
case-10 | fail→pass | 11,342 | 7,782 | -31% | 1 | 1 | 0% | 1,801 | 4,021 | +123% | 0 | 0 | — |
case-11 | fail→pass | 6,147 | 4,605 | -25% | 1 | 1 | 0% | 1,090 | 3,396 | +212% | 0 | 0 | — |
case-12 | fail→pass | 8,047 | 4,020 | -50% | 1 | 1 | 0% | 1,394 | 3,278 | +135% | 0 | 0 | — |
case-13 | fail→pass | 8,511 | 5,162 | -39% | 1 | 1 | 0% | 1,396 | 3,545 | +154% | 0 | 0 | — |
case-14 | fail→pass | 8,381 | 4,024 | -52% | 1 | 1 | 0% | 1,436 | 3,345 | +133% | 0 | 0 | — |
case-15 | fail→pass | 6,861 | 6,304 | -8% | 1 | 1 | 0% | 1,060 | 3,829 | +261% | 0 | 0 | — |
case-16 | fail→pass | 10,365 | 2,118 | -80% | 1 | 1 | 0% | 1,662 | 2,936 | +77% | 0 | 0 | — |
case-17 | fail→pass | 8,846 | 2,413 | -73% | 1 | 1 | 0% | 1,393 | 2,998 | +115% | 0 | 0 | — |
case-18 | pass→pass | 7,972 | 2,694 | -66% | 1 | 1 | 0% | 1,246 | 2,977 | +139% | 0 | 0 | — |
case-19 | fail→pass | 8,148 | 2,793 | -66% | 1 | 1 | 0% | 1,495 | 3,081 | +106% | 0 | 0 | — |
case-20 | fail→pass | 14,303 | 4,842 | -66% | 1 | 1 | 0% | 1,057 | 3,403 | +222% | 0 | 0 | — |
case-21 | fail→pass | 8,344 | 8,666 | +4% | 1 | 1 | 0% | 1,183 | 4,118 | +248% | 0 | 0 | — |
case-22 | fail→pass | 8,055 | 2,225 | -72% | 1 | 1 | 0% | 1,330 | 2,934 | +121% | 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. The headline lift of +77 percentage points is the difference between those two pass rates over the 22 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.