prepare-soil
Acerca de
Esta habilidad proporciona un método estructurado para evaluar y mejorar el suelo del jardín mediante pruebas, enmiendas, compostaje y preparados biodinámicos. Guía a los usuarios a través de pasos prácticos como pruebas en frasco, ajustes de textura y establecimiento de sistemas de compostaje. Los desarrolladores pueden utilizarla para crear funciones de planificación de jardines, análisis de salud del suelo y transición a prácticas orgánicas o de no labranza.
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/prepare-soilCopia y pega este comando en Claude Code para instalar esta habilidad
Documentación
備土
察土情並以改、堆、活活生健活之土。
用時
- 始新畦需察既土
- 水光足而植不繁(土或為因)
- 欲轉慣行為有機或生機之修
- 土已實、貧、或拒水
- 需立堆肥系
- 欲施生機備(500-508)
入
- 必要:可至所察之土(畦、田、容器)
- 可選:當前土試果(pH、N-P-K、有機質 %)
- 可選:園史(前作、所施改、栽年)
- 可選:欲種之作物或植
- 可選:法之偏(有機、生機、樸門)
法
第一步:察土
三田試無需實驗室——皆行之。
Test 1: Jar Test (Texture — Sand/Silt/Clay Ratio)
1. Fill a quart jar 1/3 full with soil from 15cm depth
2. Fill to top with water, add 1 tablespoon dish soap
3. Shake vigorously for 3 minutes, then set on level surface
4. Read layers after settling:
- Sand settles in 1 minute (bottom layer)
- Silt settles in 4-6 hours (middle layer)
- Clay settles in 24-48 hours (top layer)
5. Measure each layer as % of total soil depth
- Ideal garden soil: ~40% sand, ~40% silt, ~20% clay (loam)
Test 2: Spade Test (Structure and Compaction)
1. Push a spade into moist soil to full depth (25cm)
2. Lever up a block of soil and place on a board
3. Observe:
- Crumbles easily → good structure
- Breaks into angular blocks → compacted
- Smears or is sticky → too much clay or waterlogged
- Layers visible → hardpan or plough pan present
4. Smell the soil:
- Sweet, earthy → healthy aerobic biology
- Sour, sulphurous → anaerobic conditions (drainage problem)
Test 3: Earthworm Count (Biological Activity)
1. Dig a 30cm × 30cm × 30cm cube of soil
2. Place on a tarp or board
3. Gently break apart and count earthworms
- 0-5: Poor biology — needs organic matter
- 5-10: Fair — improving but not yet thriving
- 10-20: Good — healthy biological activity
- 20+: Excellent — this soil is alive
得:土質、構、生機之清像。罐試果、構評、蚓計皆有。
敗則:若罐試層難分,重以淨水並更激搖。若蚓計零且土酸臭,土或厭氧——改前必先治排水。
第二步:診斷與謀改
對察立改謀。
Amendment by Soil Type:
┌────────────────┬─────────────────────────┬──────────────────────────────┐
│ Diagnosis │ Symptoms │ Amendment │
├────────────────┼─────────────────────────┼──────────────────────────────┤
│ Heavy clay │ Sticky, slow drainage, │ Gypsum (calcium sulfate) │
│ │ >40% clay in jar test │ 1 kg/m², worked into top │
│ │ │ 15cm. Add coarse compost. │
│ │ │ Plant daikon radish to break │
│ │ │ hardpan biologically. │
├────────────────┼─────────────────────────┼──────────────────────────────┤
│ Sandy │ Drains instantly, won't │ Compost 5-10cm thick, worked │
│ │ hold moisture, <20% │ into top 20cm. Add biochar │
│ │ silt+clay in jar test │ (pre-charged with compost │
│ │ │ tea) for moisture retention. │
├────────────────┼─────────────────────────┼──────────────────────────────┤
│ Depleted │ Pale colour, low worm │ 10cm compost top-dress. │
│ │ count, poor growth │ Cover crop (legume mix) for │
│ │ despite watering │ nitrogen fixation. Foliar │
│ │ │ seaweed spray monthly. │
├────────────────┼─────────────────────────┼──────────────────────────────┤
│ Compacted │ Angular blocks in spade │ Broadfork (not rototiller) │
│ │ test, surface pooling, │ to fracture without │
│ │ hard when dry │ inverting. Deep mulch (15cm │
│ │ │ wood chips on paths). Plant │
│ │ │ deep-rooted comfrey. │
├────────────────┼─────────────────────────┼──────────────────────────────┤
│ Acidic (pH<6) │ Blueberries thrive but │ Wood ash (light application) │
│ │ brassicas struggle │ or dolomite lime. Test pH │
│ │ │ before and after — adjust │
│ │ │ slowly over 2 seasons. │
├────────────────┼─────────────────────────┼──────────────────────────────┤
│ Alkaline (pH>7)│ Iron chlorosis (yellow │ Elemental sulphur or acidic │
│ │ leaves, green veins) │ compost (pine needles, oak │
│ │ │ leaves). Very slow to shift. │
└────────────────┴─────────────────────────┴──────────────────────────────┘
得:合所診土情之具體改謀。
敗則:若諸情疊(如重黏且貧),先治構(gypsum + 寬叉)後生機(堆肥 + 覆作)。一時欲修諸事壓土。
第三步:建堆肥
依可用空間、料、時程擇法。
Composting Methods:
┌────────────────┬──────────────┬──────────────┬─────────────────────────┐
│ Method │ Time to │ Space Needed │ Best For │
│ │ Finished │ │ │
├────────────────┼──────────────┼──────────────┼─────────────────────────┤
│ Hot compost │ 4-8 weeks │ 1m³ minimum │ Large gardens, weed │
│ │ │ │ seed / disease kill │
├────────────────┼──────────────┼──────────────┼─────────────────────────┤
│ Cold compost │ 6-12 months │ Any size │ Low effort, small │
│ │ │ │ quantities │
├────────────────┼──────────────┼──────────────┼─────────────────────────┤
│ Vermicompost │ 3-6 months │ 0.5m² indoor │ Kitchen scraps, indoor │
│ │ │ │ / apartment gardens │
└────────────────┴──────────────┴──────────────┴─────────────────────────┘
Hot Compost Protocol:
1. Build pile in layers — 2 parts brown (carbon) to 1 part green (nitrogen)
- Brown: dried leaves, straw, cardboard, wood chips
- Green: kitchen scraps, fresh grass, manure, coffee grounds
2. Moisten each layer (damp sponge consistency)
3. Pile must be at least 1m × 1m × 1m to reach temperature
4. Internal temperature should reach 55-65°C (130-150°F) within 3-5 days
5. Turn pile when temperature drops below 45°C (every 5-7 days)
6. After 3-4 turns, cure for 2-4 weeks without turning
7. Finished compost: dark, crumbly, smells like forest floor, no recognizable inputs
Never Compost:
- Meat, dairy, oils (attract pests)
- Diseased plant material (unless hot compost reaches 60°C+ for 3 days)
- Treated wood, glossy paper
- Pet waste (pathogen risk)
得:堆肥系已立、首批進中。
敗則:若熱堆不熱:察濕(過乾或過濕)、察 C:N 比(增綠為氮)、察堆大(< 1m³ 不能穩熱)。
第四步:施不耕與覆作
護並建土構而不翻。
No-Till Sheet Mulching (New Bed from Lawn or Weeds):
1. Mow or scythe existing vegetation as low as possible
2. Layer cardboard (overlapping edges) directly on ground — no gaps
3. Wet cardboard thoroughly
4. Add 5cm compost on top of cardboard
5. Add 10-15cm organic mulch (straw, wood chips, leaves)
6. Wait 3-6 months (autumn application → spring planting)
7. Plant through mulch by pulling it aside — do not till
Cover Crop Quick Reference:
┌─────────────────┬────────────────┬───────────────────────────────┐
│ Crop │ Season │ Benefit │
├─────────────────┼────────────────┼───────────────────────────────┤
│ Crimson clover │ Autumn sow │ Nitrogen fixation, bee forage │
│ Winter rye │ Autumn sow │ Biomass, weed suppression │
│ Buckwheat │ Summer sow │ Fast cover, phosphorus mining │
│ Phacelia │ Spring/autumn │ Pollinator magnet, breaks up │
│ │ │ compaction │
│ Daikon radish │ Autumn sow │ Deep root breaks hardpan, │
│ │ │ decomposes in place over │
│ │ │ winter (bio-drill) │
└─────────────────┴────────────────┴───────────────────────────────┘
Terminate cover crops by:
- Crimp and roll (best — leaves roots in place)
- Scythe and lay as mulch
- Never rototill — this destroys the soil structure you're building
得:土終年護、生機未擾、有機質增。
敗則:若覆作未立,察播深(多需表或淺播)與濕。重播或施厚覆為替代地表覆。
第五步:生機備(可選——進階)
為循 Demeter 或生機原則之修者。
Biodynamic Preparations Overview:
┌──────┬───────────────┬──────────────────────┬─────────────────────────┐
│ Prep │ Material │ Application │ Purpose │
├──────┼───────────────┼──────────────────────┼─────────────────────────┤
│ 500 │ Horn manure │ Spray on soil, │ Stimulate soil biology, │
│ │ │ autumn & spring │ root growth, humus │
├──────┼───────────────┼──────────────────────┼─────────────────────────┤
│ 501 │ Horn silica │ Spray on foliage, │ Light metabolism, fruit │
│ │ │ morning, summer │ quality, ripening │
├──────┼───────────────┼──────────────────────┼─────────────────────────┤
│ 502 │ Yarrow │ Added to compost │ Sulphur and potassium │
│ 503 │ Chamomile │ Added to compost │ Calcium, stabilizes N │
│ 504 │ Stinging nettle│ Added to compost │ Iron, stimulates soil │
│ 505 │ Oak bark │ Added to compost │ Calcium, disease resist │
│ 506 │ Dandelion │ Added to compost │ Silica, light forces │
│ 507 │ Valerian │ Sprayed on compost │ Warmth, phosphorus │
├──────┼───────────────┼──────────────────────┼─────────────────────────┤
│ 508 │ Horsetail tea │ Spray on foliage │ Fungal disease prevent │
└──────┴───────────────┴──────────────────────┴─────────────────────────┘
Preparation 500 Application:
1. Stir 100g horn manure in 35 litres warm water
2. Stir dynamically for 1 hour — create vortex, reverse, create vortex
(alternating direction every minute)
3. Apply within 1 hour of stirring
4. Spray in large droplets on soil surface — late afternoon, descending moon
5. Apply autumn (before winter) and early spring (before sowing)
Note: Biodynamic preparations are available from certified suppliers
or local biodynamic farming groups. Making your own requires the
previous season's preparations and specific animal horn sheaths.
得:備於正時與月相施。土生機活見於 1-2 季。
敗則:若備不可得,良堆肥與覆作得 80% 之生機益。備增而非健土管之代。
第六步:癒檢點——改後察
改後六週,重察土。
Post-Amendment Soil Health Check:
1. Repeat the spade test:
- Has structure improved? (Crumbles more easily)
- Are roots penetrating deeper?
- Any remaining hardpan layers?
2. Repeat the earthworm count:
- Has the count increased? (Even 2-3 more is progress)
- Are worms distributed through the depth or just at surface?
3. Drainage test:
- Dig a 30cm hole, fill with water, let drain, refill
- Second fill should drain within 1-4 hours
- <1 hour: very free draining (may need more organic matter)
- >4 hours: still compacted or clay-heavy (continue treatment)
4. Surface observation:
- Fungal threads visible in mulch layer? (Good — decomposition active)
- Green algae on surface? (Too wet or too compacted)
- Mulch layer breaking down? (Biology is working)
Triage:
- All improving → Continue current approach, reassess next season
- Structure improved but worms low → Add more diverse organic matter
- Worms present but drainage poor → Broadfork again, add coarse material
- No improvement → Soil may have contamination — consider lab test for heavy metals
得:3 指標中至少 2 之可量進。
敗則:若 6 週無進,問題或深於表土改可治。考引入土混之高架畦為並策,續多季進地中之土。
驗
- 三田試皆完(罐、鏟、蚓)
- 自試果正診土型
- 改謀合所診情
- 堆肥系已立(熱、冷、或蚓)
- 土終年覆(覆、覆作、或活植)
- 無耕或翻土
- 改後 6 週行癒檢點
- 園誌以試果與所施改更
陷
- 不試而加:隨改費而或惡失衡。常先試
- 耕:覺有效然破土構、殺蚯蚓、引草籽至表。需鬆者用寬叉
- 裸土:露土失濕、構、生機。常覆或種覆作
- 新糞於畦:燒根並引病原。諸糞觸土前堆至少 6 月
- 不試 pH 而灰:過灰致養不可得。唯依實試果調 pH
- 求即效:建土以季年計,非週
參
cultivate-bonsai— 盆景土混(akadama/浮石/熔岩)為專土備plan-garden-calendar— 土改之時合季曆(秋灰、春堆肥)read-garden— 土觀為園察協之部heal— 改後察循癒分類模式forage-plants— 明土植關係助察野植棲make-fire— 火之木灰為傳統土改(鉀 + 灰)
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.
