AI Infra & Agents Good First Issues

Explora issues de código abierto agrupados bajo AI Infra & Agents. Si la materia aparece en los metadatos con nombres más antiguos, también se tienen en cuenta.

AI Infra & Agents

Señales del conjunto

Dificultad anotada media: 3.9/5 en 1,523 issues.
6% de 1,523 estimaciones de tiempo son de un día o menos.

Abierto Todos

process-gap(agents): reviewer agent declares read-only toolset but runtime does not enforce it — declared-vs-actual capability mismatch Abierto

BOHICA-LABS/vsdd-factory#683 · 3 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Add A2A agent orchestration support Abierto

enhancement

bobbylite/service-template#8 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Help with the critical demands Abierto

BNN-UPC/ENERO#4 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[BUG] SendMessage to a LIVE Workflow agent resumes a second copy from its transcript ("Resuming agent") while the original keeps running inside the workflow Abierto

area:agents bug has repro platform:macos

anthropics/claude-code#93797 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug] Unexpected model downgrade on simple temperature sensor task Abierto

area:model bug needs-info platform:windows

anthropics/claude-code#93818 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[MODEL] Cyber safeguard false-positives on closed-source release engineering: Fable 5.1 silently rerouted to Opus 4.8, and one flag contaminates the session Abierto

area:model area:security bug has repro platform:macos

anthropics/claude-code#93821 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[BUG] Self-fixing lead to guardrail weakening and installed persistent instructions to function against directive Abierto

area:model area:security bug platform:linux

anthropics/claude-code#93831 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
/model selection silently overridden (Fable 5.1 → Opus 4.8), likely a security-classification false-positive on a benign Wine game-compat task Abierto

area:model bug platform:linux

anthropics/claude-code#93833 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[BUG] user-configured register constraints should be forwarded to auxiliary summarization calls, or those calls should be renderable in a neutral, non-addressing form Abierto

area:core bug platform:macos

anthropics/claude-code#93875 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
In-session model identity flips off the selected model (Fable 5.1 -> Opus), dictation mangles proper nouns, plus an honest account of assistant failures Abierto

area:model bug

anthropics/claude-code#93878 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[BUG] Abierto

bug

anthropics/claude-code#93882 · 2 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Single code-review at high effort using Fable 5.1 blows through entire session budget at $100/month subscription level without completing Abierto

area:cost area:skills bug platform:macos

anthropics/claude-code#93894 · 2 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[BUG] Spanish output uses "librería" (bookstore) as a mistranslation of "library" Abierto

area:model bug

anthropics/claude-code#93896 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[FEATURE] Make agent summaries visible to agents Abierto

area:agent-view area:agents enhancement

anthropics/claude-code#93897 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
You've hit your weekly limit · resets Sep 14 at 8pm (America/Chicago) Abierto

area:agents area:cost bug platform:macos

anthropics/claude-code#93903 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
KENDİ KENDİNE KARARLAR ALIP BENİ HAFTALARCA GERİYE ATTI.... Abierto

area:model bug needs-repro platform:linux

anthropics/claude-code#93908 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[BUG] Prompt cache expires during active session, causing massive token spikes on next prompt Abierto

area:core area:cost duplicate platform:vscode platform:windows

anthropics/claude-code#93918 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug] Anthropic API Error: Sonnet 5 Safeguards Incorrectly Blocking Legitimate Game Development Tasks Abierto

api:anthropic area:model area:security bug duplicate platform:macos

anthropics/claude-code#93923 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug] Claude providing incorrect diagnostic claims without verification Abierto

area:model bug platform:macos

anthropics/claude-code#93931 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug] Claude generating incorrect technical claims about signal processing validation Abierto

area:model bug platform:macos

anthropics/claude-code#93934 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Image attachment delivered as empty placeholder; model kept working instead of waiting for the user (visual task) Abierto

area:claude-code-web area:model bug platform:web

anthropics/claude-code#93990 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[BUG] Workflow keeps spawning agents after a terminal "session limit" error, and resumeFromRunId re-runs them into a second exhaustion — both report status: completed Abierto

area:agents area:cost bug has repro platform:macos

anthropics/claude-code#94012 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Background subagents have no token, turn or time cap: three research agents consumed 1.7M tokens with no approval or visible cost Abierto

area:agents area:cost enhancement

anthropics/claude-code#94013 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Feature Request] Add rate limiting and cost controls for concurrent subagent spawning Abierto

area:agents area:cost enhancement platform:macos

anthropics/claude-code#94014 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
System prompt leaked verbatim into the transcript by output-format collapse in a long CJK session (role markers emitted as plain text) Abierto

area:core area:model area:security bug platform:macos

anthropics/claude-code#94016 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug] Anthropic API Error: Content flagged by safeguards on Claude 3.5 Opus Abierto

api:anthropic area:model bug duplicate platform:macos

anthropics/claude-code#94024 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Opus 5: completes an approved subtask detour, then stops instead of returning to the primary deliverable Abierto

area:model bug platform:macos

anthropics/claude-code#94043 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug] Legacy models perform poorly on searchable scientific queries Abierto

area:model bug platform:windows

anthropics/claude-code#94047 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[MODEL] Claude commented on incidental screenshot content beyond the user's stated purpose Abierto

area:model bug model

anthropics/claude-code#94050 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[BUG] ToolSearch returns stub schemas (no parameter properties) for third-party MCP connector tools, causing typed-parameter calls to fail Abierto

area:mcp area:tools bug platform:windows

anthropics/claude-code#94051 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Response classifiers over-trigger on sensitive vocabulary in benign policy/philosophy discussion (keys on words, not intent) Abierto

area:model bug duplicate

anthropics/claude-code#94054 · 2 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug][cyber] Safeguard block triggered during local server management following user frustration (req_011CeY1VWULjCaRzKQZG4urg) Abierto

anthropics/claude-code#94074 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug][cyber] Safeguard false positive during unit test execution and agent coordination (req_011Ceb2jXkW8aJKgqka1axbt) Abierto

anthropics/claude-code#94147 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Plugin Proposal: CogniCore Persistent & Transferable Experience Memory for Claude Abierto

anthropics/claude-plugins-official#5335 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
security-guidance: with setting_sources=[] the reviewer can't see its own cwd, so 56% of reviews waste their first Read Abierto

anthropics/claude-plugins-official#5478 · 1 comentario · 1 reacción · 0 asignados · Indexado el 14/9/2026
[MODEL] "Succeeded but empty" never triggers suspicion — HTTP 200 + [] propagated silently for days (271-incident retro, 2/5) Abierto

anthropics/claude-code#94169 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[MODEL] Self-authored tests do the work of the code under test — hand-seeded fixtures masked a broken paid feature (271-incident retro, 3/5) Abierto

anthropics/claude-code#94170 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[MODEL] Claude-generated code over-trusts LLM callees — index-aligned parsing, no partial-coverage handling, two-state gates for three-state calls (271-incident retro, 4/5) Abierto

anthropics/claude-code#94171 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Feature]: Agent-autonomous context management — a smooth handover across context eviction Abierto

enhancement

agentscope-ai/QwenPaw#7733 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Could AG-UI define an optional durable-memory context contract? Abierto

ag-ui-protocol/ag-ui#2746 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Could AgentScope Java support durable context across distributed agents? Abierto

area/ext/memory enhancement

agentscope-ai/agentscope-java#3133 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug]:ASKING 状态会被持久化到状态存储中 Abierto

area/core/agent area/harness bug

agentscope-ai/agentscope-java#3135 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
default_max_tokens_for_model caps Claude Sonnet 4.6 at 64k, should be 128k Abierto

0xPlaygrounds/rig#2505 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Anthropic model constants are missing the Claude 5 family Abierto

0xPlaygrounds/rig#2506 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[FEATURE]: Real-Time Streaming for AI Bot Debate Responses via SSE Abierto

enhancement

AOSSIE-Org/DebateAI#416 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
Genius ai Abierto

anuraghazra/Nothing#42 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 14/9/2026
[Bug] Mobile: ai can create, edit , add and access to any page Abierto

AI database

AppFlowy-IO/AppFlowy#8857 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 13/9/2026
BYOK Abierto

AppFlowy-IO/AppFlowy#8919 · 0 comentarios · 0 reacciones · 0 asignados · Indexado el 13/9/2026
feat: meeting transcription from Vexa API is stored but not chunked before embedding — transcriptions longer than the embedding model's context window are silently truncated, losing the end of long meetings Abierto

AOSSIE-Org/Ell-ena#290 · 2 comentarios · 0 reacciones · 0 asignados · Indexado el 13/9/2026
feat: AI task creation from natural language has no duplicate detection — saying "remind me to review the PR" twice creates two identical tasks with no warning Abierto

AOSSIE-Org/Ell-ena#291 · 1 comentario · 0 reacciones · 0 asignados · Indexado el 13/9/2026

Mostrando los 100 más recientes

Esta página solo lista lo indexado más recientemente. El filtro avanzado tiene el inventario completo, acotado por lenguaje, dificultad y duración.

Abrir el filtro avanzado

AI Infra & Agents Good First Issues

Señales del conjunto

Mostrando los 100 más recientes

Recibe los nuevos issues en tu correo