grade-tcg-card
Acerca de
Esta habilidad califica programáticamente tarjetas coleccionables utilizando los estándares de PSA, BGS o CGC, analizando centrado, superficie, bordes y esquinas. Es compatible con múltiples juegos de cartas y está diseñada para evaluaciones previas al envío, valoración de colecciones y disputas sobre el estado. El proceso incluye una evaluación basada en observación y proporciona una calificación final con un intervalo de confianza.
Instalación rápida
Claude Code
Recomendadonpx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/grade-tcg-cardCopia y pega este comando en Claude Code para instalar esta habilidad
Documentación
Grade TCG Card
Assess + grade trading card, professional grading standards (PSA, BGS, CGC). Observation-first protocol from meditate skill prevents grade anchoring — most common grading bias.
When Use
- Evaluate card before professional grading service submission
- Pre-screen collection for high-grade candidates worth submitting
- Settle disputes about card condition between buyers + sellers
- Learn to grade consistent with structured assessment protocol
- Estimate grade-dependent value spread for specific card
Inputs
- Required: Card identification (set, number, name, variant/edition)
- Required: Card images or physical description (front + back)
- Required: Grading standard to apply (PSA 1-10, BGS 1-10 with subgrades, CGC 1-10)
- Optional: Known market value at different grades (grade-value analysis)
- Optional: Card game (Pokemon, Magic: The Gathering, Flesh and Blood, Kayou)
Steps
Step 1: Clear Bias — Observation Without Prejudgment
From meditate Step 2-3: observe card without anchoring to expected grade or market value.
- Set aside any knowledge of card market value
- Do NOT look up recent sales or population reports before grading
- Know card is "valuable"? Acknowledge bias explicit:
- "I know this card is worth $X in PSA 10. I am setting that aside."
- Examine card as physical object first, not collectible
- Note initial gut impression but do NOT let it anchor assessment
- Label premature grade thoughts as "anchoring", return to observation
Got: Neutral starting state. Card assessed purely on physical condition, not market expectations. Grade anchoring (knowing value before grading) = #1 source of grading inconsistency.
If fail: Bias sticky (high-value card makes you want to see 10)? Write bias down explicit. Externalizing reduces influence. Proceed only when can examine card as physical object.
Step 2: Centering Assessment
Measure card print centering on both faces.
- Measure border width all four sides of front face:
- Left vs right border (horizontal centering)
- Top vs bottom border (vertical centering)
- Express as ratio: 55/45 left-right, 60/40 top-bottom
- Repeat for back face
- Apply grading standard centering thresholds:
PSA Centering Thresholds:
+-------+-------------------+-------------------+
| Grade | Front (max) | Back (max) |
+-------+-------------------+-------------------+
| 10 | 55/45 or better | 75/25 or better |
| 9 | 60/40 or better | 90/10 or better |
| 8 | 65/35 or better | 90/10 or better |
| 7 | 70/30 or better | 90/10 or better |
+-------+-------------------+-------------------+
BGS Centering Subgrade:
+------+-------------------+-------------------+
| Sub | Front (max) | Back (max) |
+------+-------------------+-------------------+
| 10 | 50/50 perfect | 50/50 perfect |
| 9.5 | 55/45 or better | 60/40 or better |
| 9 | 60/40 or better | 65/35 or better |
| 8.5 | 65/35 or better | 70/30 or better |
+------+-------------------+-------------------+
- Record centering score each axis + applicable subgrade
Got: Numeric centering ratios for both faces with corresponding grade/subgrade identified. Most objective measurement in grading process.
If fail: Borders too narrow to measure (full-art, borderless prints)? Note "centering N/A — borderless", skip to Step 3. Some grading services apply different standards for borderless cards.
Step 3: Surface Analysis
Examine card surface for defects.
- Examine front surface under good lighting:
- Print defects: ink spots, missing ink, print lines, color inconsistency
- Surface scratches: visible under direct + angled light
- Whitening on surface: haze or clouding of surface layer
- Indentations or impressions: dents visible under raking light
- Staining or discoloration: yellowing, water marks, chemical damage
- Examine back surface same criteria
- Check factory defects vs handling damage:
- Factory: print lines, miscut, crimping — may be less penalized
- Handling: scratches, dents, stains — always penalized
- Rate surface condition:
- Pristine (10): flawless under magnification
- Near-pristine (9-9.5): minor imperfections only under magnification
- Excellent (8-8.5): minor wear visible to naked eye
- Good (6-7): moderate wear, multiple minor defects
- Fair or below (1-5): significant damage visible
Got: Detailed surface inventory with each defect located, described, severity-rated. Factory vs handling defects distinguished.
If fail: Images too low-resolution for surface analysis? Note limitation, provide grade range not point grade. Recommend physical inspection.
Step 4: Edge + Corner Evaluation
Assess card edges + corners for wear.
- Examine all four edges:
- Whitening: white spots or lines along colored edges (most common defect)
- Chipping: small pieces of edge layer missing
- Roughness: edge feels uneven or has micro-tears
- Foil separation: on holofoil cards, check delamination at edges
- Examine all four corners:
- Sharpness: corner tip crisp + pointed
- Rounding: corner tip worn to curve (slight, moderate, heavy)
- Splitting: layer separation visible at corner (dings)
- Bending: corner turned or creased
- Rate edge + corner condition same scale as surface
- Note which specific corners/edges have worst condition
Got: Per-edge + per-corner condition assessment. Worst individual corner/edge typically limits overall grade.
If fail: Card in sleeve or toploader obscures edges? Note which areas couldn't be fully assessed.
Step 5: Assign Final Grade
Combine sub-assessments into final grade.
- For PSA grading (single number 1-10):
- Final grade limited by weakest sub-assessment
- Card with perfect surface but 65/35 centering caps at PSA 8
- Apply "lowest limits" principle, adjust up if other areas exceptional
- For BGS grading (four subgrades → overall):
- Assign subgrades: Centering, Edges, Corners, Surface (each 1-10 in 0.5 steps)
- Overall = weighted average, but lowest subgrade limits overall
- BGS 10 Pristine needs all four subgrades at 10
- BGS 9.5 Gem Mint needs average 9.5+ with no subgrade below 9
- For CGC grading (similar to PSA with subgrades on label):
- Assign Centering, Surface, Edges, Corners
- Overall follows CGC proprietary weighting
- State final grade with confidence:
- "PSA 8 (confident)" — clear grade, unlikely higher or lower
- "PSA 8-9 (borderline)" — could go either way at grading service
- "PSA 7-8 (uncertain)" — limited assessment data
Got: Final grade with confidence level. BGS → all four subgrades reported. Grade supported by evidence from Steps 2-4.
If fail: Assessment inconclusive (can't tell if surface mark scratch or dirt)? Provide grade range, recommend professional grading. Never assign confident grade with insufficient data.
Checks
- Bias check completed before grading (no grade anchoring)
- Centering measured on both faces with ratios recorded
- Surface examined for scratches, print defects, staining, indentations
- All four edges + corners individually assessed
- Factory vs handling defects distinguished
- Final grade supported by evidence from each sub-assessment
- Confidence level stated (confident, borderline, uncertain)
- Grading standard correctly applied (PSA/BGS/CGC thresholds)
Pitfalls
- Grade anchoring: Knowing card value before grading biases toward "hoped-for" grade. Always assess physically first
- Ignoring the back: Back surface + back centering count. Many graders over-focus on front
- Confusing factory with handling defects: Factory print line different from scratch, but both affect grade
- Over-grading holofoils: Holographic + foil cards hide surface scratches until viewed at right angle. Use multiple light angles
- Centering optical illusions: Art placement can make centering appear better or worse than is. Measure borders, not art
See Also
build-tcg-deck— Deck building where card condition affects tournament legalitymanage-tcg-collection— Collection management with grade-based valuationmeditate— Source of observation-without-prejudgment technique adapted for grading bias prevention
Repositorio GitHub
Habilidades relacionadas
llamaguard
OtroLlamaGuard es el modelo de Meta de 7-8B parámetros para moderar las entradas y salidas de LLM en seis categorías de seguridad como violencia y discurso de odio. Ofrece una precisión del 94-95% y puede implementarse usando vLLM, Hugging Face o Amazon SageMaker. Utiliza esta skill para integrar fácilmente filtrado de contenido y barreras de seguridad en tus aplicaciones de IA.
cost-optimization
OtroEsta Skill de Claude ayuda a los desarrolladores a optimizar los costes en la nube mediante el ajuste de tamaño de recursos, estrategias de etiquetado y análisis de gastos. Proporciona un marco para reducir los gastos en la nube e implementar una gobernanza de costes en AWS, Azure y GCP. Úsala cuando necesites analizar los costes de infraestructura, ajustar el tamaño de los recursos o cumplir con restricciones presupuestarias.
quantizing-models-bitsandbytes
OtroEsta habilidad cuantiza LLMs a precisión de 8 o 4 bits utilizando bitsandbytes, logrando una reducción de memoria del 50-75% con pérdida mínima de precisión. Es ideal para ejecutar modelos más grandes en memoria GPU limitada o para acelerar la inferencia, admitiendo formatos como INT8, NF4 y FP4. La habilidad se integra con HuggingFace Transformers y permite entrenamiento QLoRA y optimizadores de 8 bits.
dispatching-parallel-agents
OtroEsta Skill de Claude despliega múltiples agentes para investigar y solucionar 3 o más problemas independientes de forma concurrente. Está diseñada para escenarios que involucran fallos no relacionados que pueden resolverse sin estado compartido o dependencias. Su capacidad principal es la resolución paralela de problemas, asignando un agente por cada dominio problemático independiente para maximizar la eficiencia.
