🔧 AI Nachrichten Major AI platforms go down in unprecedented simultaneous outage(03.09.2026 um 17:34 Uhr)
🔧 AI Nachrichten ChatGPT, Claude, and Grok Down? Users Report Widespread Outages(03.09.2026 um 19:14 Uhr)
🔧 AI Nachrichten OpenAI Launches GPT-6 Astra, Says We May Have Entered the AGI Era(03.09.2026 um 22:08 Uhr)
🔧 AI Nachrichten Claude Comes to CarPlay as Fifth Major AI Chatbot App(05.09.2026 um 05:31 Uhr)
🔧 AI Nachrichten OpenAI’s GPT-6 Astra Is AGI, Says NVIDIA CEO Jensen Huang(07.09.2026 um 06:31 Uhr)
🔧 AI Nachrichten Blame AI companies for Mac mini and Mac Studio shortage(31.08.2026 um 10:32 Uhr)
🔧 AI Nachrichten Major AI platforms go down in unprecedented simultaneous outage(03.09.2026 um 17:34 Uhr)
🔧 AI Nachrichten ChatGPT, Claude, and Grok Down? Users Report Widespread Outages(03.09.2026 um 19:14 Uhr)
🔧 AI Nachrichten OpenAI Launches GPT-6 Astra, Says We May Have Entered the AGI Era(03.09.2026 um 22:08 Uhr)
🔧 AI Nachrichten Claude Comes to CarPlay as Fifth Major AI Chatbot App(05.09.2026 um 05:31 Uhr)
🔧 AI Nachrichten OpenAI’s GPT-6 Astra Is AGI, Says NVIDIA CEO Jensen Huang(07.09.2026 um 06:31 Uhr)
🔧 AI Nachrichten Blame AI companies for Mac mini and Mac Studio shortage(31.08.2026 um 10:32 Uhr)

💾 Tools 🕛 kürzlich 9 Min Lesezeit
0

GitHub: v17.2.11

↗ Quelle (GitHub · github.com)
🗣️ Stimme:
📑 Inhaltsübersicht
🐙
$ git clone https://github.com/can1357/oh-my-pi.git

@oh-my-pi/pi-ai


Breaking Changes



  • Fixed handling of GitHub Copilot's model_not_available_for_integrator error to prevent unnecessary retries, preserving the actionable available models list.


Added



  • Added support for reporting Cursor personal monthly USD quotas and remaining balances, labeled by verified profile email accounts.


Fixed



  • Fixed an issue where ANTHROPIC_BASE_URL was ignored for Anthropic chat requests, ensuring requests are routed to the configured host and forwarding ANTHROPIC_CUSTOM_HEADERS to non-official gateways.

  • Fixed an issue where a legacy pre-organization login credential could persist and cause a permanent error row in omp usage even after a successful organization-scoped re-login.

  • Fixed an issue where lazy provider streams (including Amazon Bedrock, Google, Cursor, Devin, and Ollama) ignored model-specific idle timeouts, which previously caused healthy but slow reasoning turns to prematurely time out.

  • Improved error classification for Simplified Chinese quota-exhaustion and rate-limit messages, ensuring affected credentials are correctly rotated or backed off instead of being treated as unknown errors.

  • Classified subscription and plan-cap 429 responses as rotatable usage limits rather than transient rate-limit throttles, enabling smoother credential rotation.


@oh-my-pi/pi-catalog


Fixed



  • Increased the default stream idle-timeout floor on Amazon Bedrock to 900 seconds for reasoning and adaptive-thinking models (such as Claude) to prevent premature watchdog timeouts during long reasoning stretches.

  • Fixed Devin model families (including SWE-1.7, Claude 5, Gemini 3.6 Flash, Kimi K3, Grok 4.5, and Inkling) to correctly group as logical models with reasoning-effort routing instead of separate wire variants.

  • Added missing context-window and output-token limits for dynamically discovered Alibaba Token Plan models.


@oh-my-pi/pi-coding-agent


Added



  • Added support for the Agent Plugins 1.0.0 standard, enabling automatic discovery, validation, and secure execution of compliant plugin packages.

  • Added the omp share <session> command to share saved sessions by ID prefix or file path without launching the agent.

  • Added the AGENT=1 environment variable to child processes spawned by coding-agent to allow downstream tools to detect agent-driven execution.


Changed



  • Consolidated Exa web-search configuration under exa.enabled, automatically migrating legacy exa.enableSearch values and removing obsolete Researcher and Websets settings.

  • Removed stale computer.backend values during configuration migration.

  • Updated documentation and error messages for the JavaScript/TypeScript debug adapter (js-debug-adapter) to clarify supported installation paths (Mason, standalone tarball, or JS_DEBUG_DAP_SERVER).


Fixed



  • Fixed an issue where /reload-plugins and the Agent Control Center failed to propagate updated agent definitions to existing tools without a restart.

  • Fixed legacy Pi extensions failing to load when calling pi.unregisterProvider(), ensuring provider replacements take effect immediately.

  • Fixed zero-width daemon readiness and wait regex matches being rejected by the hub wire decoder.

  • Fixed proxy model discovery preferring bundled catalog names over proxy-reported names, allowing omp models refresh to correctly update display names.

  • Fixed Windows compiled binary builds failing due to backslash-separated paths in Bun.Glob.scan producing invalid JavaScript in virtual modules.

  • Fixed the Ctrl+O (app.tools.expand) shortcut not expanding truncated tool output when a tool-approval prompt or selection dialog had keyboard focus.

  • Improved omp commit error reporting when pre-commit or commit-msg hooks fail, displaying the hook's own message and exiting non-zero cleanly instead of printing bundled source code.

  • Fixed omp commit --push exiting with code 0 without pushing when the working tree is already clean; it now correctly pushes existing commits.

  • Fixed omp commit exiting with code 0 when the commit agent failed and fell back to a mechanical commit; it now exits non-zero to indicate the fallback was used.

  • Fixed strict output schemas being rejected when native JSON Schema definition maps contain ref or applicator branches use properties without type.

  • Fixed shell syntax extraction in cd <path> && ... commands to prevent redirects, extra arguments, or shell expansions from being incorrectly absorbed into the structured working directory path.

  • Applied reason-specific backoff to transient rate-limit retries and consolidated exhausted retry errors.

  • Fixed session-tree rows rendering as empty bullets for bookkeeping entries (such as title changes, credential pins, and mode changes); these are now hidden by default and properly labeled in all mode.

  • Fixed extension and custom tools inheriting same-named built-in TUI renderers, which could overwrite successful results with incorrect status text.

  • Fixed prewalk lifecycle handling to prevent plan injection on rejected same-model/same-effort arms, ensure consumed plan nudges do not return after context rebuilds, and prevent settings-enabled prewalk from implicitly re-arming restored sessions.

  • Fixed the todo completion reminder interrupting pauses when waiting for non-English questions (such as Chinese, Japanese, Korean, or Spanish prompts ending in or ?).

  • Normalized resolved file paths in read summaries, PDF image handles, and notebook errors to prevent agents from learning malformed paths.

  • Fixed a bug where a per-turn before_agent_start system prompt override was silently dropped during base-prompt rebuilds.

  • Fixed ACP session/load and session/resume failing with ACP session not found for sessions created under the legacy hashed project-directory scheme by falling back to a global ID scan.

  • Fixed vault://<name>?op=... commands targeting the active vault instead of the named vault in Obsidian CLI queries.

  • Fixed the status-line session_name segment to honor the statusLine.sessionAccent setting, falling back to the theme's accent color when disabled.

  • Fixed automatic agent.continue() paths failing to run context-fit maintenance when reverting to a smaller-context model after a cooldown expiry.

  • Fixed /handoff reporting "Handoff cancelled" for actual generation or stream timeout errors, ensuring the real error is surfaced.


@oh-my-pi/hashline


Changed



  • Pasting an empty named register (PUT … @name with no matching capture) now surfaces a warning listing available registers and removes the span target instead of throwing an error.


Fixed



  • Fixed an issue where pipe-numbered read/search rows copied into top-level and bare-body patch payloads were not properly recovered ( in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in made their first contribution in made their first contribution in made their first contribution in made their first contribution in made their first contribution in made their first contribution in made their first contribution in made their first contribution in made their first contribution in

Vollständiger Original-Bericht
Ausführliche Details, Code-Beispiele & Hersteller-Stellungnahme auf github.com.
↗ Original-Artikel auf github.com lesen
Wie bewertest du diesen Beitrag?
1 Klick Feedback
Teilen mit Netzwerk & Team:

Community-Analysen & Experten-Meinungen 0

Verfasse deine eigene Analyse, teile Workarounds oder diskutiere diesen Vorfall im Blog.
Noch keine Community-Analyse verfasst. Markiere einen Textabschnitt oder klicke oben auf Eigene Analyse verfassen“!
Community Pulse: Relevanz-Einschätzung
1 Klick Experten-Votum
🔴 Akute Relevanz 55%
🟡 In Evaluierung 28%
🟢 Keine Auswirkung 12%
Spannende Innovation 5%
Verwandte Story-Cluster & Quellen (Vektor-KI)
Port 8095 Engine
3 Quellen
GPT-6 Astra Release Today? OpenAI’s Next Major AI Model Is Almost Here
1 Quelle
Apple accuses OpenAI of destroying evidence as trade-secrets fight intensifies
1 Quelle
Major AI platforms go down in unprecedented simultaneous outage