github can1357/oh-my-pi v18.2.10
on GitHub

4 hours ago

@oh-my-pi/pi-agent-core

Fixed

  • Fixed stream finalization when a provider ends without emitting a completion or error event, ensuring the final assistant message is preserved and corresponding message lifecycle events are emitted.
  • Fixed tool execution being incorrectly skipped when host steering callbacks reject during a tool batch.
  • Fixed stream hangs and preserved the original error when host aside-commit or discard callbacks fail.

@oh-my-pi/pi-ai

Added

  • Added Claude saved-reset discovery and redemption, including session-only resets, grant eligibility, expiry, and safe retry handling.

Fixed

  • Fixed custom OpenAI-compatible extension streamers failing when no compatibility configuration was provided.
  • Fixed local provider sign-in for LM Studio, llama.cpp, and vLLM so an empty API key is not treated as successful authentication.
  • Fixed Vercel AI Gateway models backed by non-Anthropic providers failing tool calls because of strict schema validation; requests now retry with compatible non-strict tool handling.
  • Improved AWS Bedrock authentication by refreshing expired AWS SSO sessions automatically, reducing the need to run aws sso login again.
  • Fixed Bedrock tool-enabled requests when tool descriptions are included in the system prompt.
  • Improved Alibaba Token Plan (Beijing) quota reporting across workspaces and made gateway rejection codes visible in error logs.
  • Fixed the tool-call loop guard so repeated identical calls continue to be redirected after the detection threshold is reached.
  • Fixed valid required null values inside tool argument unions being removed before dispatch (#12523 by @cswenor).
  • Signing in to a local provider (lm-studio, llama.cpp, vllm) with an empty key paste no longer reports the provider as logged in while its requests go out unauthenticated. (#12436 by @xiechimon)
  • Fixed every turn failing with 400 Invalid schema for function '<tool>' … Missing '<param>' on Vercel AI Gateway models served from a non-Anthropic upstream (e.g. openai/gpt-5.6-sol): the translated strict-tool rejection now triggers the existing non-strict retry instead of failing the turn (#12760 by @primitive-type).
  • Expired AWS SSO access tokens are now refreshed via the SSO OIDC refresh_token grant instead of failing with sso-token-expired, so Bedrock profiles keep working between aws sso login runs the same way the AWS CLI does (#12736 by @nwbb).
  • Fixed Bedrock rejecting tool-enabled requests when tool descriptions are inlined into the system prompt (#12732 by @mustafaabidali).
  • Alibaba Token Plan (Beijing) quota reporting no longer pins requests to a single workspace, and HTTP-200 gateway rejections now log their error code (#12395 by @Dante-dan).
  • The tool-call loop guard keeps redirecting when a model continues the same identical call past the detection threshold instead of firing only once (#12709 by @F0Rextasy).
  • Fixed OpenAI-compatible Gemini gateways losing message-level thought signatures when replaying tool-call history.
  • Added support for explicitly disabling reasoning with reasoning_effort: "none" through Chat Completions authentication gateways.
  • Improved Bedrock resilience by retrying transient in-stream internal server, service unavailable, and throttling errors.
  • Fixed Anthropic prompt-cache keep-alive refreshes for turns that used forced tool selection.
  • Fixed auth-broker usage reports incorrectly sharing usage limits between Team members with shared workspace and organization identifiers.
  • Fixed OpenAI Codex requests hanging when an error response body is delayed.
  • Fixed Kimi usage reporting so monthly totals and code quotas are shown alongside the five-hour usage window.
  • Bedrock no longer sends provider-invalid payloads when an errored tool result contains an image; the image is hoisted into a sibling block (#12865 by @roboomp).
  • Gemini, Vertex, and Cloud Code Assist requests no longer include the unsupported minP/repetitionPenalty sampling fields, which caused 400s when set globally (#12850 by @roboomp).

@oh-my-pi/pi-catalog

Changed

  • Updated default models for anthropic, amazon-bedrock, cloudflare-ai-gateway, kilo, litellm, opencode-zen, vercel-ai-gateway, and zenmux to claude-opus-5-5
  • Updated default model for commandcode to claude-sonnet-5

Fixed

  • Local OpenAI-compatible backends—including llama.cpp, LM Studio, vLLM, Ollama, and custom loopback or LAN hosts—now support maxTokens values above 64k without incorrectly applying the hosted OpenAI limit.
  • Fixed Union Alpha requests on OpenCode Go and Zen by using the correct Messages API.
  • Updated SuperGrok’s Grok 4.7 metadata to expose its documented 500K context window and low, medium, high, and xhigh thinking levels.
  • Fixed GPT-OSS tool interactions on Google Antigravity and Gemini CLI so function-call history is preserved correctly.
  • Fixed Devin model discovery for Enterprise credentials by falling back to the legacy Windsurf catalog when native discovery does not return the full model list.
  • Curated Xiaomi Token Plan (China) MiMo V2.6 metadata: context/output limits, reasoning, and image input (#12841 by @roboomp).

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

Added

  • Added live benchmark results table with real-time model ranking and per-kind performance metrics
  • Added dedicated prefill throughput reporting for prefill-focused benchmarks
  • Added /record slash command to capture terminal sessions as replayable .ompcast files
  • Added omp play CLI for terminal-based playback of session recordings
  • Added intent descriptions to judgment batching
  • Added live progress tracking for judgment batches in the TUI
  • Added Claude saved resets to usage views and /usage reset, with automatic blocked-limit recovery and expiring-reset redemption controlled by claudeResets.
  • Added support for searching embedded harness documentation with find and omp find using omp:// scopes, including file-specific searches and :start-end selectors; results open directly through canonical omp:// URLs.

Changed

  • Refined AI-assisted git staging verification to reduce false positives
  • Updated omp bench default profile to chat and improved CLI flag documentation
  • Coalesced judgment batch drain operations for better performance under high load
  • Updated server-side fallback documentation and logic to target claude-opus-5-5
  • Added support for claude-opus-5-5 to model priority registry
  • Updated the read tool guidance to decode images inline by default and require an explicit :img selector for SVG rendering.
  • Improved model discovery and fallback behavior: authentication failures are surfaced in the /models hub, and models without a matching role-specific fallback now use the default fallback chain.
  • Improved resilience for subagents by retrying provider stream failures that occur after partial output and preserving configured ordered model fallbacks at startup.
  • MCP OAuth with Google issuers now requests offline access so refresh tokens can be issued; repeated auth-broker token rotations also preserve the required refresh and client metadata.
  • MCP servers from omp-plugins now expand ${CLAUDE_PLUGIN_ROOT} and ${OMP_PLUGIN_ROOT} in commands, arguments, and working directories.
  • /review now uses the session's current working directory after /move or /wt.
  • Pasted and dragged image files now retain their original filesystem paths so the agent can act on the source files directly.
  • Custom sessions can now be moved across filesystems without losing transcripts or artifacts.
  • hub jobs now returns a compact, non-consuming status summary instead of replaying completed output or consuming pending auto-delivery.
  • The display-reset shortcut now works while the ask dialog has keyboard focus, and tab.press() provides a clear error for the legacy argument order.
  • Wayland keyboard input now follows the compositor's active XKB layout instead of assuming a US layout.
  • LSP diagnostics now refresh when watched files are created or deleted and after a server reload.
  • Compiled bytecode binaries now start correctly when bundled dependencies use import.meta.resolve.

Fixed

  • Fixed JavaScript eval assignments in cells containing top-level await so they persist into subsequent cells.
  • Fixed skill hints becoming out of sync with the active prompt after discarded rebuilds and in advisor sessions.
  • Restored pi.pi.askToolRenderer for extensions that replace the built-in ask tool, preserving native rendering.
  • Fixed npm plugin upgrades and reinstalls leaving stale or duplicate manifest entries that could break bun install.
  • Fixed eval waits longer than approximately 24.8 days returning immediately because of native timer overflow.
  • Fixed deleted sessions being resurrected from stale rewrite backups.
  • Fixed /collab relay connections honoring HTTPS_PROXY and NO_PROXY.
  • Fixed sessions remaining blocked by queued turns or Hindsight auto-recall after disposal or cancellation.
  • Fixed edits to auto-generated files aborting the entire turn; they now return a tool-scoped error.
  • Fixed Edit handling of invalid overlapping selections in multibyte text so the worker reports a match error instead of panicking.
  • Fixed local memory consolidation on case-insensitive filesystems when project path casing changes between launches.
  • Fixed first-time Xcode MCP connections on macOS by allowing the signed omp binary to request Apple Events permission.
  • Fixed stale or duplicated TTSR trigger events during streaming.
  • Fixed MCP OAuth credentials retaining their refresh endpoint and client metadata across repeated token rotations.
  • Fixed local model and provider retry behavior for streamed and partially buffered failures.
  • Fixed memory and session cleanup issues that could leave stale artifacts or inconsistent state.
  • lsp.formatOnWrite now prefers a dedicated isLinter formatter server when a type-checker also claims the file (#12847 by @roboomp).
  • /extensions no longer shows OMP-installed marketplace capabilities as disabled behind the foreign-plugin opt-in gate (#12849 by @roboomp).

Removed

  • Removed support for image query parameters (?q=) and bare image paths in the read tool.
  • Custom models now honor provider-level transport: pi-native and send requests to the native gateway (#12845 by @joshrzemien).
  • Fixed live models that match no retry.fallbackChains role primary (e.g. Fable after /model) resolving no chain, so a wait longer than retry.maxDelayMs aborted the session instead of walking default (#12421).
  • Fixed skill hints drifting from the active prompt after discarded rebuilds or in advisor sessions (#12148 by @jerome-benoit).
  • Restored askToolRenderer on the extension namespace (pi.pi.askToolRenderer) after the pi-tui renderer migration dropped it, so extensions that shadow the built-in ask tool can keep the native rendering again. (#12694 by @xiechimon)
  • Model discovery rejected with 401/403 now surfaces an authentication error in the /models hub instead of a silently empty model list. (#12436 by @xiechimon)
  • Pasted or dragged image files now reach the agent with their original filesystem path, so it can read and act on the source file directly; clipboard screenshots keep working unchanged. (#12404 by @xiechimon)
  • /review now runs VCS operations against the live session cwd after /move or /wt instead of the session-start checkout (#12712 by @F0Rextasy).
  • Reinstalling or upgrading an npm plugin no longer leaves stale or duplicate manifest edges that broke bun install (#12727 by @F0Rextasy).
  • TTSR now emits one ttsr_triggered event per streamed violation instead of one per evaluation pass (#12729 by @F0Rextasy).
  • Eval wait() timeouts above ~24.8 days no longer overflow the native timer and return immediately (#12731 by @F0Rextasy).
  • MCP OAuth against Google issuers now requests access_type=offline so refresh tokens are issued (#12737 by @F0Rextasy).
  • Deleting a session now also removes its stale .bak rewrite backups so the picker cannot resurrect it (#12746 by @F0Rextasy).
  • /collab relay WebSockets now honor HTTPS_PROXY/NO_PROXY like other transports (#12762 by @jacobcolyvan).
  • tab.press() now rejects the inverted press(selector, key) call with a hint naming the corrected (key, { selector }) form, instead of the key parser's opaque Unknown key: <selector> (#12136) (#12266 by @danilouchoa).
  • The display-reset shortcut (app.display.reset, alt+l by default) now fires while the ask dialog holds keyboard focus, instead of being dropped silently; the #11215 global-listener promotion covered the other four editor display actions but missed this one (#12217) (#12262 by @danilouchoa).
  • Custom sessions can move across filesystems without losing their transcript or artifacts (#12360, #12378 by @Dante-dan).
  • Fixed hub jobs replaying full output for every settled job and consuming pending auto-delivery; it now returns a compact non-consuming status summary (#12547 by @pedropaulovc).
  • Compiled bytecode binaries now start correctly when bundled dependencies use import.meta.resolve (#12133 by @andrebrait).
  • Subagents now retry provider stream errors that arrive after buffered partial output, and such failures are reported as transport errors instead of schema-invalid results (#12752 by @bse-ai).
  • Edits targeting auto-generated files now return a tool-scoped rejection instead of aborting the whole turn (#12499 by @Dante-dan).
  • Subagents with an ordered model fallback keep it reachable on startup when the parent default role shares the same primary model (#12377 by @Dante-dan).
  • omp-plugins MCP servers now substitute ${CLAUDE_PLUGIN_ROOT}/${OMP_PLUGIN_ROOT} in command, args, and cwd (#12801 by @holny).

@oh-my-pi/pi-natives

Fixed

  • Fixed stale workspace addons failing when first used; the error now identifies the addon and provides the command to rebuild native addons.
  • Fixed background bash jobs hanging indefinitely when output forwarding stalls.

@oh-my-pi/snapcompact

Fixed

  • Fixed summaries incorrectly advertising a ¶think: scope when thinking content is excluded from the archive.

@oh-my-pi/omp-stats

Fixed

  • Fixed background statistics spans to use unique identifiers and close jobs correctly.

@oh-my-pi/pi-tui

Added

  • Added Claude and Codex saved-reset counts, availability, and expiry to usage views, with provider-specific confirmation and uniquely identified account options.
  • Added terminal detection and capability support for otty and rio, including Kitty graphics and true-color support where available, plus otty hyperlinks and notifications.
  • Added the public stripTerminalSequences export for extensions that need to remove terminal control sequences.

Changed

  • Added support for multiple concurrent TUI paint listeners to enable simultaneous session recording and streaming
  • Coalesced status event updates for progress-based operations to reduce TUI render overhead

Fixed

  • Composer shape previews now use the full available overlay width instead of being clipped at 96 columns.
  • Fixed cursor placement when recalling prompts from history, keeping single-line entries at the end and preserving the appropriate position for multi-line entries.
  • Restored modified-key handling and terminal notifications over SSH sessions running inside tmux.
  • Fixed typed Enter occasionally being interpreted as a literal newline when terminal input events are batched.

What's Changed

  • singularityapi login flow by @H4vC in #12707
  • singularityapi main platform by @H4vC in #12740
  • fix(catalog): singularityapi image transport and GPT-5.6 reasoning ladder by @H4vC in #12751
  • feat(find): search omp:// harness docs via materialized corpus by @H4vC in #12758
  • fix(retry): walk default fallbackChains when live model matches no role (#12421) by @shah1git in #12620
  • fix(compat): model side-turn capability policy by @gokceneraslan in #12629
  • feat(catalog): split singularityapi into dev and tech providers by @H4vC in #12803
  • fix(task): preserve subagent retry fallback role on startup by @Dante-dan in #12377
  • fix(hub): summarize jobs without consuming output by @pedropaulovc in #12547
  • fix(agent): append zero-event trailing message to context and emit lifecycle events by @xiechimon in #12605
  • fix(tui): detect the otty terminal with Kitty graphics and placeholder support by @xiechimon in #12665
  • fix(snapcompact): gate ¶think: legend on includeThinking by @F0Rextasy in #12703
  • fix(setup): run link-omp natively on Windows by @F0Rextasy in #12704
  • fix(ai): re-detect continued identical tool calls past the threshold by @F0Rextasy in #12709
  • fix(changelog): merge duplicate 18.0.7 sections by @F0Rextasy in #12717
  • fix(catalog): curate xai-oauth grok-4.7 with documented thinking ladder by @eggpeat in #12725
  • fix(stats): ensure unique background span ids and correct job closing by @404MaximWang in #12726
  • fix(plugins): prune stale npm edge before bun install by @F0Rextasy in #12727
  • fix(ttsr): emit one ttsr_triggered per violation, not per evaluation by @F0Rextasy in #12729
  • fix(eval): chunk oversized wait timers past the 32-bit limit by @F0Rextasy in #12731
  • fix(ai): omit empty Bedrock tool descriptions instead of sending "" by @mustafaabidali in #12732
  • fix(ai): refresh expired AWS SSO tokens instead of erroring by @nwbb in #12736
  • fix(mcp): request access_type=offline from Google issuers by @F0Rextasy in #12737
  • fix(session): delete stale .bak siblings with the session by @F0Rextasy in #12746
  • fix(coding-agent): retry provider stream errors in subagents and stop reporting them as schema failures by @bse-ai in #12752
  • fix(ai): classify gateway-translated strict tool schema rejections by @primitive-type in #12760
  • fix(collab): honour relay WebSocket proxies by @jacobcolyvan in #12762
  • fix(catalog): enable supportsFunctionPartId for gpt-oss on google-antigravity by @tuandinh0801 in #12771
  • fix(ai): map Bedrock stream exception shapes to their service-model status by @nick-maderight in #12775
  • fix(ai): accept reasoning_effort none on Chat Completions auth-gateway by @holny in #12789
  • fix(tui): add stripTerminalSequences by @RockinPaul in #12795
  • fix(coding-agent): substitute CLAUDE/OMP plugin root in omp-plugins MCP servers by @holny in #12801
  • Document browser MCP filtering by browser.enabled by @pavel-kalmykov in #12804
  • fix(ai): preserve message-level Gemini thought signatures by @roboomp in #10631
  • test(coding-agent): isolate agent directory in advisor toggle tests by @andrebrait in #12130
  • fix(build): retain ESM semantics in compiled bytecode by @andrebrait in #12133
  • fix(skills): freeze provider-side skill hint across mid-session skillful toggles by @jerome-benoit in #12148
  • fix(coding-agent): keep display reset working while the ask dialog holds focus by @danilouchoa in #12262
  • feat(tui): detect rio terminal with kitty graphics and true color by @danilouchoa in #12263
  • fix(browser): reject inverted tab.press(selector, key) with an actionable hint by @danilouchoa in #12266
  • fix(ai): avoid pinning Beijing quota to one workspace by @Dante-dan in #12395
  • fix(catalog): honor maxTokens on local OpenAI hosts by @ton-anywhere in #12433
  • fix(coding-agent): keep generated-file errors tool-scoped by @Dante-dan in #12499
  • fix(natives): reported a stale workspace addon instead of " is not a function" by @iacore in #12698
  • fix(commands): review uses live session cwd after /move or /wt by @F0Rextasy in #12712
  • fix(tui): use full overlay width for composer shape preview by @F0Rextasy in #12713
  • fix(catalog): route Union Alpha through OpenCode Messages API by @Dante-dan in #12380
  • fix(coding-agent): carry the original filesystem path of pasted images to the agent by @xiechimon in #12404
  • fix(tui): open recalled prompt history at a consistent caret by @H4vC in #12814
  • feat(collab): publish whether a host session is working by @alphastorm in #12844
  • fix(tui): resolve terminal identity through tmux by @roboomp in #12810
  • fix(tui): stop raw-paste classifier from eating stall-batched enter by @roboomp in #12811
  • fix(pi-edit): reject overlapping sloppy selection spans instead of panicking by @roboomp in #12817
  • fix(auth): preserve mcp oauth refresh material by @roboomp in #12820
  • fix(ai): prevented auth-broker Team usage collisions by @roboomp in #12821
  • fix(macos): added apple events signing entitlement by @roboomp in #12823
  • fix(memories): fold cwd casing into memory scope keys on win32 by @roboomp in #12826
  • fix(ai): drop forced tool_choice from zero-output cache refresh by @roboomp in #12827
  • fix(ai): bound codex error-body reads with the pre-response deadline by @roboomp in #12831
  • fix(shell): bound chunk-forward stall so a wedged consumer cannot strand a background bash job by @roboomp in #12832
  • fix(hindsight): cancel in-flight auto-recall on abort by @roboomp in #12833
  • fix(agent): guard trailing checkSteering in runTool against rejection by @roboomp in #12813
  • fix(agent): isolate aside-commit / discardAsides host callbacks by @roboomp in #12815
  • fix(coding-agent): suspend loop on Esc even mid-iteration by @H4vC in #12851
  • fix(catalog): restored legacy Devin discovery by @roboomp in #12858
  • fix(ai): parsed Kimi monthly usage quotas by @roboomp in #12855
  • fix(extensions): bind ExtensionAPI methods by @roboomp in #12856
  • fix(lsp): refresh diagnostics after workspace changes by @roboomp in #12857
  • fix(ai): preserve required nested nulls in tool argument unions by @cswenor in #12523
  • fix(catalog): curate xiaomi mimo v2.6 metadata by @roboomp in #12841
  • fix(coding-agent): preserve provider transport on custom models by @joshrzemien in #12845
  • fix(lsp): prefer formatter servers for formatOnWrite by @roboomp in #12847
  • fix(extensions): exempt omp-origin marketplace plugins from dashboard opt-in gate by @roboomp in #12849
  • fix(ai): drop Gemini-unsupported minP/repetitionPenalty from google transports by @roboomp in #12850
  • test(coding-agent): flush the deferred prewarm tick before restoring mocks by @andrebrait in #12863
  • fix(bedrock): hoist images out of error toolResults by @roboomp in #12865
  • fix(ai): probe claude resets when usage lists null program blocks by @roboomp in #12886
  • perf: faster extension loading and off-thread highlighter warmup by @H4vC in #12908
  • fix(ai): send per-message effort when an Anthropic session starts on the API default by @H4vC in #12909

New Contributors

Full Changelog: v18.2.8...v18.2.10

Check out latest releases or
releases around can1357/oh-my-pi v18.2.10

Don't miss a new oh-my-pi release

NewReleases is sending notifications on new releases.

Get notifications