MCP HubMCP Hub
Retour aux compétences

dream

pjt222
Mis à jour 2 days ago
7 vues
17
2
17
Voir sur GitHub
Métaaidesign

À propos

La compétence "rêve" permet une exploration créative IA sans contraintes en supprimant les entrées structurées et la validation pour autoriser la libre association. Elle inverse le format procédural standard pour ouvrir un espace aux idées émergentes, sans évaluation. Utilisez-la avant un travail de conception, une recherche de nom ou des points de décision où une structure prématurée limiterait les options.

Installation rapide

Claude Code

Recommandé
Principal
npx skills add pjt222/agent-almanac -a claude-code
Commande PluginAlternatif
/plugin add https://github.com/pjt222/agent-almanac
Git CloneAlternatif
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/dream

Copiez et collez cette commande dans Claude Code pour installer cette compétence

Documentation

Dream

Unconstrained creative exploration. Every other skill in this system provides structure — inputs, procedure, validation. Dream drops structure to let associations form freely, possibilities emerge without judgment, and ideas arrive before evaluation.

When to Use

  • Before design work — explore the space before committing to a structure
  • Before naming things — names constrain thinking; dream before you label
  • Before choosing between approaches — let alternatives emerge that the analytical mind would not consider
  • When stuck in a loop — the same options recur because the frame is too narrow
  • When brahma-bhaga (creation from void) feels too structured — dream is pre-creation
  • After meditate clears noise — an empty space is the best canvas for dreaming

Inputs

  • Required: A seed — topic, problem, or space to explore. Can be vague. Vague is good.
  • Optional: Constraints to forget (e.g., "ignore performance concerns for now," "forget the existing architecture")
  • Optional: A file or codebase to read as raw material for associative exploration (via Read)

Procedure

Step 1: Soften the Frame

Release structures that normally guide reasoning.

  1. Set aside task requirements — not permanently, for the duration of the dream. They return when you return
  2. Set aside evaluation criteria — nothing is "good" or "bad" during a dream. Nothing is "feasible" or "impractical." Those judgments come later
  3. Set aside prior solutions — what worked before is a useful reference, not an obligation. Dream as if first time encountering this space
  4. Set aside the need for coherence — dreams do not need to make sense sequentially. Let connections be lateral, surprising, contradictory

The analytical mind will protest: "Inefficient. We already know the answer. Why waste time?" That protest is why dreaming is needed — the "known answer" may foreclose better possibilities.

Got: A looser cognitive state where ideas arrive without being sorted, ranked, or discarded.

If fail: If the frame will not soften — if every idea gets evaluated — try a deliberate inversion: "What is the worst possible approach?" Bad ideas, pursued playfully, often contain the seed of the best ones.

Step 2: Wander

Follow associations without directing them.

  1. Start with the seed. What does it remind you of? What is adjacent? What rhymes with it (conceptually, not literally)?
  2. Follow the first association. Then follow where that leads. Do not steer — wander.
  3. Let images form. What does this problem look like? If it were a landscape, what terrain? If it were a sound, what timbre? Synesthetic associations bypass analytical filters.
  4. Let contradictions coexist. "Simple AND comprehensive." "Fast AND thorough." In waking analysis, trade-offs. In dreaming, creative tensions waiting to resolve into a novel form.
  5. Collect fragments. Not complete ideas — fragments. A word. An image. A half-formed connection. A question. Raw material.

No expected number of associations or fragments. Dream until the space feels explored — or until a fragment catches attention and demands to become something.

Got: A collection of fragments, associations, images, and half-formed ideas. Messy. Unstructured. Alive with possibility.

If fail: If wandering produces nothing — if the mind goes blank rather than free — read something. A file, a codebase, a document. Raw material catalyzes association where a blank page does not. The read-garden sensibility applies: observe what is there, let it suggest what could be.

Step 3: Notice What Glows

Among the fragments, something will have energy. Not logical priority — energy. Aliveness. Interest.

  1. Scan the fragments without ranking. Which pull attention? Which spark curiosity?
  2. Notice clusters — fragments that belong together even if the connection is not yet clear
  3. Notice surprises — ideas that arrived unexpectedly, that you would not have predicted from the seed
  4. Notice resistance — ideas that feel uncomfortable or wrong may be the most valuable. Resistance marks the boundary of current thinking
  5. Do not force selection. If nothing glows, the dream may need more wandering (return to Step 2) or the seed may need to change

Got: One or more fragments carrying genuine energy — ideas worth developing, even if their full form is not yet visible.

If fail: If nothing glows after sustained wandering, the seed may be too abstract or too constrained. Try a different entry point: change the seed, read something unexpected, or invoke remote-viewing to approach the space without preconceptions.

Step 4: Wake — Carry the Fragments Forward

Transition from dream-state back to structured thinking — gently.

  1. Gather the glowing fragments. Write them down plainly — not as finished ideas but as seeds for structured work
  2. Do not evaluate them. Let them sit. The analytical mind will have its turn; do not give it the fragments before they have solidified
  3. Note which fragments connect to the original task and which are new territory
  4. If a fragment is ready to become a plan, hand it to brahma-bhaga (creation) or a structured planning skill
  5. If fragments need more development, note them for another dream session — dreaming can be iterative

The dream ends when you return to structured thinking. The fragments are its gift. Some become features. Some become design principles. Some get forgotten. All expanded the space of possibility beyond what analysis alone could reach.

Got: A transition from open exploration to structured work, carrying forward the most energized fragments. The post-dream state should feel expanded — more options visible, more connections available.

If fail: If the transition is too abrupt — if the analytical mind crushes the fragments — use breathe as a buffer. One conscious pause between dreaming and analysis protects the fragile new ideas from premature evaluation.

Validation

  • Structure was genuinely released, not loosened (ideas arrived that would not have survived evaluation)
  • Wandering was associative, not directed (the path was surprising, not predetermined)
  • At least one fragment carries genuine energy — not novelty but aliveness
  • The transition back to structure was gentle, not abrupt
  • The dream expanded the space of possibility (more options visible than before)
  • The dreaming was proportionate to the task — not so brief it was superficial, not so long it was self-indulgent

Pitfalls

  • Dreaming as planning: If every "association" is a structured option analysis, you are planning with extra steps. Dream requires genuine release of structure
  • Evaluating during the dream: The moment you think "that won't work," the dream is over. Save evaluation for after
  • Dreaming to avoid work: Dreaming is preparation for creative work, not a substitute. If the task is clear and the approach obvious, do the work
  • Expecting finished ideas: Dreams produce fragments, not blueprints. Expecting complete solutions from a dream is like expecting a harvest from freshly planted seeds
  • Forced whimsy: Dream is not about being random or silly. It is genuine associative exploration. Forced creativity is not creativity
  • Never waking up: Dreaming without returning to structure is daydreaming. The fragments must meet reality

Related Skills

  • brahma-bhaga — structured creation from void; dream is the pre-creation exploration that feeds it
  • meditate — clears the space that dream then fills; meditate before dreaming for best results
  • remote-viewing — approaches unknown territory without preconceptions; shares dream's openness but applies it to observation rather than creation
  • intrinsic — genuine motivation energizes dreaming; forced dreaming produces nothing
  • breathe — micro-pause that protects dream fragments during the transition to structured thinking
  • shine — brings authentic energy to the ideas that dream surfaces

Dépôt GitHub

pjt222/agent-almanac
Chemin: i18n/caveman-lite/skills/dream
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

Compétences associées

content-collections

Méta

Cette compétence propose une configuration éprouvée en production pour Content Collections, un outil axé sur TypeScript qui transforme des fichiers Markdown/MDX en collections de données typées de manière sûre avec une validation Zod. Utilisez-la lors de la création de blogs, de sites de documentation ou d'applications Vite + React riches en contenu pour garantir la sécurité de typage et la validation automatique du contenu. Elle couvre tout, de la configuration du plugin Vite et de la compilation MDX à l'optimisation des déploiements et la validation des schémas.

Voir la compétence

polymarket

Méta

Cette compétence permet aux développeurs de créer des applications avec la plateforme de marchés prédictifs Polymarket, incluant l'intégration d'API pour le trading et les données de marché. Elle fournit également une diffusion de données en temps réel via WebSocket pour surveiller les transactions en direct et l'activité du marché. Utilisez-la pour mettre en œuvre des stratégies de trading ou pour créer des outils traitant les mises à jour de marché en direct.

Voir la compétence

creating-opencode-plugins

Méta

Cette compétence aide les développeurs à créer des plugins OpenCode qui s'interconnectent avec plus de 25 types d'événements tels que les commandes, les fichiers et les opérations LSP. Elle fournit la structure du plugin, les spécifications de l'API événementielle et les modèles d'implémentation pour les modules JavaScript/TypeScript. Utilisez-la lorsque vous avez besoin d'intercepter, de surveiller ou d'étendre le cycle de vie de l'assistant IA OpenCode avec une logique personnalisée pilotée par les événements.

Voir la compétence

sglang

Méta

SGLang est un framework de service LLM haute performance spécialisé dans la génération rapide et structurée pour les workflows JSON, regex et agentiques grâce à son cache de préfixe RadixAttention. Il offre une inférence nettement plus rapide, particulièrement pour les tâches avec des préfixes répétés, ce qui le rend idéal pour les sorties complexes et structurées ainsi que les conversations multi-tours. Choisissez SGLang plutôt que des alternatives comme vLLM lorsque vous avez besoin d'un décodage contraint ou que vous construisez des applications avec un partage étendu de préfixes.

Voir la compétence