issues Search Results · language:Dune language:Python language:Python language:HTML language:JavaScript language:JavaScript
Filter by
47M results
HISTORIA DE USUARIO
Como admin requiero que el usuario genere una cuenta en la aplicación para acceder a la misma
CRITERIOS DE ACEPTACIÓN
ESCENAARIO1. DATOS VÁLIDOS
DADO un formulario de registro CUANDO ...
uliseScript
What
Assemble agent system prompt from 9 ordered blocks: platform-instructions, agent-identity, agent-personality (AGENT.md
body), memory, governance-knowledge, skills (eager), available-skills (lazy ...
context
feat
maddhruv
id: d8094b03b42a ts: 2026-06-02T16:00
• 全戦術共通の1日取引上限に達した: count=251 ≥ 200(hard block) いやいや今日は一件も発火してないけど？
user-instruction
yuusakuichio-hash
What
Configure embedding provider in llm.config.yaml embeddings: section. Priority: OpenAI text-embedding-3-small (cloud) →
Ollama local (no external key) → MongoDB $text lexical fallback. Dimension-change ...
feat
llm
What
POST /api/v1/chat accepts optional model param (specific model ID or tier name). Override applies to that turn only; not
persisted to conversation or agent config.
Why
Lets users experiment with ...
feat
llm
What
api_key_ref in LLM config is a logical secret key resolved via SecretsService.resolve(key). env://VAR escape hatch reads
from env var directly. Credentials cached after first resolution.
Why
LLM ...
feat
llm
What
soul/llm.config.yaml defines tiers, providers, model IDs, and credential refs. Validated by TypeBox/AJV at startup.
Missing required tier or provider causes non-zero exit.
Why
LLM config is operator-controlled. ...
feat
llm
What
Each tier has an ordered list of providers. FallbackProvider walks the list on failure. Hard errors (auth failure, model
not found) abort without fallback. Transient errors (rate limit, timeout) ...
feat
llm
What
Three model tiers: quick (haiku-4-5), standard (sonnet-4-6), complex (opus-4-8). auto tier selects based on rules:
x-llm-decision → complex; autonomy map (full→quick, supervised→standard, approval-required→complex); ...
feat
llm
What
Wire Vercel AI SDK LanguageModelV4 as the universal LLM adapter. Support @ai-sdk/* packages for 55+ providers plus
openai-compatible base_url path for self-hosted models.
Why
Provider-agnostic ...
feat
llm
Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip! Restrict your search to the title by using the in:title qualifier.
Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip! Restrict your search to the title by using the in:title qualifier.