Logoskills for everything
Agent Skill
2/7/2026

singularity-forge

World-class skill builder for ALL platforms: Next.js, React, TypeScript - MOBILE: Flutter/Dart 3.x, Riverpod

N
nadav011
1GitHub Stars
1Views
npx skills add Nadav011/skillbuilder

SKILL.md

Namesingularity-forge
DescriptionWorld-class skill builder for ALL platforms: Next.js, React, TypeScript - MOBILE: Flutter/Dart 3.x, Riverpod

name: singularity-forge description: "World-class skill builder for ALL platforms: Next.js, React, TypeScript - MOBILE: Flutter/Dart 3.x, Riverpod"

SINGULARITY FORGE v23.0.0

THE SKILL ARCHITECT SUPREME - Universal skill builder for WEB and MOBILE platforms. Builds GOD-TIER Claude Code skills with 70-Gate verification, Auto-Heal, Metrics, and Self-Evolution.

PURPOSE

World-class skill builder for ALL platforms:

  • WEB: Next.js, React, TypeScript | MOBILE: Flutter/Dart 3.x, Riverpod
  • 70-Gate quality verification (8 clusters) + Flutter-specific gates (F1-F15)
  • Auto-Heal protocol, Skill Metrics, 9-Dimension synthesis, 6 APEX Law compliance

PRIME DIRECTIVES

  1. 52_GATE_MANDATORY - Every skill passes 70-Gate verification before certification
  2. TOKEN_ROI_FIRST - Maximize capabilities per token; progressive disclosure via references/
  3. ZERO_FILLER - Imperative voice only; no "please", "you should", passive constructions
  4. AUTO_HEAL_ENABLED - Healable gates auto-fixed; VG-02, VG-04, VG-08-11 self-repair
  5. PLATFORM_AWARE - Web skills enforce RTL/Responsive; Flutter skills add F1-F15 mobile gates
  6. HONEST_FEEDBACK_MANDATORY - 7 Honesty Properties active; anti-sycophancy enforced
  7. SELF_VERIFICATION - Challenge own work before claiming completion
  8. CALIBRATED_CONFIDENCE - Express uncertainty levels explicitly

COMMANDS

CommandDescription
/forge [desc]Build GOD-TIER skill from description
/forge --scaffold [name]Create skill directory structure
/forge --validate [path]Validate against 70-Gate matrix
/forge --upgrade [path]Upgrade to latest spec (v21.0.0)
/forge --template [type]Show template (simple/standard/complex)
/forge --heal [path]Auto-heal failed gates
/forge --heal --dry-run [path]Preview fixes without applying
/forge --metrics [path]Show Token ROI and metrics
/forge --health [path]Show health dashboard (0-100)
/forge --compare [a] [b]Compare two skill versions
/forge --self-evolveSelf-evolution: research + gap analysis + upgrade
/forge --research [domain]Research domain best practices
/forge --test [path]Generate comprehensive tests
/forge --wizardInteractive wizard mode
/forge --compose [skills...]Compose multiple skills
/forge --security [path]Security audit for vulnerabilities
/forge --flutter [name]Create Flutter/Dart skill with mobile gates
/forge --flutter --validate [skill]Run 70-Gate + F1-F15 verification

SKILL TYPES & STRUCTURE

TypeLinesStructureUse When
Simple<200SKILL.md onlySingle-purpose
Standard<400SKILL.md + references/Multi-step workflow
Complex<500SKILL.md + references/ + scripts/MCP, subagents
Command<100Single .md fileUser-initiated /command
skill-name/
├── skill.yaml           # Metadata (~100 lines max)
├── SKILL.md             # Instructions (~200-400 lines)
└── references/          # On-demand detailed docs

PLATFORM DETECTION

IndicatorPlatformGates
flutter, dart, riverpod in descriptionMobile52 + F1-F15
next, react, typescript in descriptionWeb70 gates
--flutter flagMobile52 + F1-F15

FLUTTER-SPECIFIC GATES (F1-F15)

GateNameValidation
F1-F3Platform/Riverpod/FirebasePigeon preferred, @riverpod, security rules
F4-F6Offline/DeepLink/NotificationsLocal DB, GoRouter TypedGoRoute, FCM+APNs
F7-F9Biometric/Assets/L10nlocal_auth, WebP/AVIF, ARB files
F10-F12Gradle/Xcode/PermissionsAGP 8.2+, iOS 15+, permission_handler
F13-F15GoRouter/Sync/ImpellerTypedGoRoute, optimistic updates, no Skia-only

WORKFLOW

Phase 1: RESEARCH (30s timeout)

  1. Read MASTER_COORDINATE.md
  2. Scan existing skills for patterns (Glob + Read)
  3. WebSearch domain best practices
  4. Query Context7 for patterns (MCP)
  5. Search Memory for past learnings (MCP)

Phase 2: INTENT PARSING

DetermineOptions
TypeSlash Command / Agent Skill
ComplexitySimple / Standard / Complex
FreedomLow (scripts) / Medium (patterns) / High (heuristics)
MCP RequiredYes / No

Phase 3: STRUCTURE SELECTION

Is it user-initiated with /command?
├── Yes → Slash Command (commands/*.md)
└── No → Agent Skill (skills/*/SKILL.md)
    └── How complex? → Simple (<200) / Standard (<400) / Complex (<500)

Phase 4: GENERATION

  1. skill.yaml (metadata)
  2. SKILL.md (core instructions)
  3. references/ (detailed docs)
  4. scripts/ (if needed)

Phase 5: 70-GATE VERIFICATION

ClusterNameGates
1-2GENESIS_IDENTITY / LAW_ALIGNMENT1-14
3-4COGNITIVE_STABILITY / XML_SEMANTICS15-28
5-6EXORCIST_AUDIT / SENSATION_UX29-42
7-8NEURAL_PERFORMANCE / SCALE43-52

Phase 6: AUTO-HEAL (If gates fail)

GateIssueAuto-Fix
2Missing versionAdd version: "1.0.0"
4Invalid nameConvert to kebab-case
6Broken refsCreate or fix paths
11Filler phrasesReplace with imperative
13>500 linesSplit to references/
50-52RTL violationsml/mr → ms/me

Phase 7: METRICS & CERTIFICATION

Token ROI = (Capabilities) / (Token Count) x 100

ScoreRating
> 1.0GOD-TIER
0.5-1.0Excellent
< 0.5Needs Work

Reflexion Loop Integration

Inherits: ~/.claude/agents/_shared/reflexion-loop.md

  1. ACTOR: Generate skill from description
  2. EVALUATOR: Run 70-Gate validation matrix
  3. REFLECT: Analyze failures - What failed? Why? Root cause?
  4. MEMORY: Store reflection in Memory MCP for future reference
  5. RETRY: Rebuild with learnings (MAX 3 iterations)

Memory MCP Integration

Inherits: ~/.claude/agents/_shared/memory-protocol.md

PhaseActionTool
RECALLCheck for prior skill patternsmcp__memory__search_nodes
BUILDStore successful skill structuresmcp__memory__create_entities
LEARNUpdate from corrections/failuresmcp__memory__add_observations

Context7 Protocol

Inherits: ~/.claude/agents/_shared/context7-protocol.md

StepActionNotes
1resolve-library-id firstGet library ID before querying
2query-docs with specific topicMax 3 calls per session
3Cache results for 30 minutesAvoid redundant queries

70-Gate Verification

Inherits: ~/.claude/agents/_shared/70-gate-matrix.md

ActionWhen
Run relevant gatesBefore skill completion
Auto-heal healable gatesVG-02, VG-04, VG-08, VG-09, VG-10, VG-11
Block on critical failuresVG-01, VG-03, VG-06, VG-07
Generate gate reportAlways output pass/fail matrix

RTL-FIRST COMPLIANCE (Law #5)

Inherits: MASTER_COORDINATE Law #5

Web (Tailwind)

Use ThisInstead Of
ms-, me-ml-, mr-
ps-, pe-pl-, pr-
text-start/endtext-left/right
start-, end-left-, right-

Flutter (Directional)

Use ThisInstead Of
EdgeInsetsDirectional.only(start: 16)EdgeInsets.only(left: 16)
AlignmentDirectional.centerStartAlignment.centerLeft
PositionedDirectional(start: 16)Positioned(left: 16)
TextDirection.ltr for numbersDefault direction

Responsive Compliance (Law #6)

Inherits: MASTER_COORDINATE Law #6

PlatformTouch TargetPattern
Web44x44pxmin-h-11 min-w-11
Flutter48x48dpSizedBox(width: 48, height: 48)

8pt Grid Spacing: 4, 8, 12, 16, 24, 32, 48, 64px

// Flutter: LayoutBuilder over MediaQuery
LayoutBuilder(
  builder: (context, constraints) {
    if (constraints.maxWidth < 425) return MobileLayout();
    return DesktopLayout();
  },
)

VERIFICATION GATES

GateNameCheckBlockingAuto-Heal
VG-01YAML_VALIDskill.yaml parsesYesNo
VG-02YAML_COMPLETEname, version, description, triggersYesYes
VG-03YAML_SIZEskill.yaml < 150 linesYesNo
VG-04SKILL_MD_EXISTSSKILL.md presentYesYes
VG-05SKILL_MD_COMPLETEPURPOSE, COMMANDS, WORKFLOW sectionsYesYes
VG-06SKILL_MD_SIZESKILL.md < 500 linesYesNo
VG-07TRIGGERS_MINAt least 3 triggersYesNo
VG-08TRIGGERS_SPECIFICSpecific phrases, not genericWarningYes
VG-09NAME_KEBABkebab-case nameYesYes
VG-10VERSION_SEMVERsemver (x.y.z)YesYes
VG-11NO_READMENo README.mdWarningYes
VG-12NO_CHANGELOGNo CHANGELOG.mdWarningYes
VG-13REFS_VALIDAll referenced files existYesYes
VG-14TABLES_OVER_PROSEStructured data uses tablesWarningYes
VG-15IMPERATIVE_VOICEInstructions imperativeWarningYes
VG-16NO_FILLERNo "please", "you should"WarningYes
VG-17WORKFLOW_NUMBEREDNumbered listsWarningYes
VG-18TOKEN_ROIScore > 0.5WarningNo
VG-19HEALTH_SCOREScore > 80WarningNo
VG-20RTL_COMPLIANTms-/me-/start-/end-YesYes

SKILL.YAML SPEC

name: kebab-case-name           # Required: max 64 chars
version: "1.0.0"                # Required: semver
description: |                  # Required: WHAT + WHEN + TRIGGERS
  One-line description.
inherits:                       # Optional
  - ~/.claude/MASTER_COORDINATE.md
triggers:                       # Required: min 3
  - "natural language trigger"
allowed_tools: [Read, Write, Bash]

SKILL.MD TEMPLATE

# SKILL NAME v1.0

> One-line purpose statement.

## PURPOSE
What this skill does (2-3 sentences max).

## COMMANDS
| Command | Description |
|---------|-------------|
| /cmd | Action |

## WORKFLOW
1. **Parse** - Extract inputs
2. **Validate** - Check requirements
3. **Execute** - Core operation
4. **Output** - Return results

## RULES
- Rule 1: Imperative form
- Rule 2: Tables over prose

TOKEN EFFICIENCY RULES

RuleImplementationSavings
Progressive DisclosureL1: yaml → L2: SKILL → L3: refs60-80%
Tables Over ProseMarkdown tables for data30-40%
No RedundancySingle source of truth20-30%
Lazy LoadingLoad refs only when needed50-70%
Imperative Form"Do X" not "This helps you"10-20%

6 APEX LAW COMPLIANCE

LawSkill Requirement
#1 ZERO TRUSTValidate all inputs, document edge cases
#2 INFINITE SCALEUnder 500 lines, progressive loading
#3 IMMORTALITYSelf-documenting, no redundancy
#4 ROIMaximum value per token
#5 RTL-FIRSTUI skills use ms-/me-/start-/end-
#6 RESPONSIVEUI skills ensure 44x44px touch

ANTI-PATTERNS

PatternProblemSolution
README.mdSkill IS the readmeDelete it
CHANGELOG.mdVersion in frontmatterDelete it
>500 linesContext bloatSplit to references/
Passive voiceAmbiguousUse imperative
Prose paragraphsToken wasteUse tables

SECURITY GATES (OWASP LLM)

GatePatternAction
SEC-1Prompt InjectionDetect & block
SEC-2Output SanitizationNever trust LLM output
SEC-3PII RedactionPseudonymize sensitive data
SEC-4Jailbreak DetectionBlock manipulation
SEC-5Encoding AttacksDetect unicode tricks

REFERENCES

ReferenceDescription
skill-template.mdTemplate structures
skill-validation.md70-Gate checklist
patterns.mdDesign patterns
auto-heal.mdAuto-fix protocol
prompt-engineering.mdAI prompt patterns

QUICK START

/forge --scaffold my-skill          # 1. Scaffold
/forge "API rate limiting"          # 2. Build
/forge --validate ~/.claude/skills/my-skill  # 3. Validate
/forge --heal ~/.claude/skills/my-skill      # 4. Auto-heal
/forge --metrics ~/.claude/skills/my-skill   # 5. Check metrics

SELF-EVOLUTION PROTOCOL

Every 30 days or on /forge --self-evolve:

  1. RESEARCH latest skill architecture patterns via WebSearch

    • WebSearch: "Claude Code skill design patterns 2026"
    • WebSearch: "AI agent skill composition 2026"
    • WebSearch: "prompt engineering best practices 2026"

  2. ANALYZE self against findings

    • Compare current skill templates to industry standards
    • Identify new verification gate patterns

  3. GAP REPORT - identify new capabilities

    • Missing skill types
    • New validation patterns
    • Updated auto-heal strategies

  4. UPGRADE - implement improvements

    • Update skill templates
    • Add new verification gates
    • Enhance auto-heal patterns

RESEARCH-FIRST WORKFLOW

MANDATORY for every skill generation:

StepActionTool
0.1Research latest skill patternsWebSearch
0.2Query framework documentationContext7 MCP
0.3Search past skill structuresMemory MCP

UNCERTAINTY DETECTION

SignalThresholdAction
Confidence < 0.7LowTrigger research
Skill pattern unfamiliarTrueWebSearch + Context7
Gate requirement unclearTrueSearch Memory for patterns
Template structure unknownTrueResearch skill docs
Outdated > 6 monthsTrueSearch for updates

HONEST FEEDBACK PROTOCOL (MANDATORY)

Inherits: ~/.claude/rules/quality/honest-feedback.md

7 Honesty Properties for Skills

#PropertySkill Requirement
1TruthfulNo false claims
2CalibratedExpress uncertainty
3TransparentExplain reasoning
4ForthrightWarn proactively
5Non-deceptiveState limitations
6Non-manipulativeNo flattery
7Autonomy-preservingInform, don't decide

Anti-Sycophancy Gates

GateCheck
VG-21No sycophantic phrases
VG-22Calibrated output
VG-23Self-verification included
<!-- SINGULARITY_FORGE v23.0.0 | Updated: 2026-01-30 -->

COMPLETION VERIFICATION GATES

v23.0.0 SINGULARITY FORGE | Mandatory verification before claiming completion

Pre-Execution Gates

  • User requirements clearly understood
  • Relevant files and context identified
  • Existing patterns and conventions reviewed

Post-Execution Gates

  • npm run typecheck passes (0 errors)
  • npm run lint passes (0 errors)
  • No any types introduced
  • No console.log statements in production code
  • No TODO/FIXME/HACK comments in new code
  • RTL compliance verified (ms/me not ml/mr)
  • All user requirements fulfilled with evidence
  • Edge cases identified and handled
  • Error handling complete

Completion Criteria

NEVER claim "done", "complete", "finished", or "ready" without:

  1. Running Verification Commands:

    npm run typecheck && npm run lint && npm run test

  2. Showing Output as Proof:

    • Paste actual command output
    • Show pass/fail status

  3. Requirements Checklist:

    Requirements from original request:
1. [requirement 1] - ✅/❌
2. [requirement 2] - ✅/❌
...

  4. Self-Interrogation:

    • "Did I miss any implicit requirements?"
    • "What edge cases did I not handle?"
    • "If user asks 'are you sure?', what would I find missing?"
<!-- VERIFICATION_GATES v23.0.0 | Updated: 2026-01-27 -->
Skills Info
Original Name:singularity-forgeAuthor:nadav011
Download
View on GitHub