github can1357/oh-my-pi v15.10.12
on GitHub

4 hours ago

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

Added

  • Added AgentLoopConfig.getDisableReasoning so callers can override disableReasoning per LLM call, mirroring getReasoning.
  • Added transformProviderContext to AgentOptions/AgentLoopConfig: an optional hook applied to the assembled provider context after conversion, normalization, and append-only handling, but before telemetry capture and provider send.

Fixed

  • Fixed Agent runs so explicit reasoning disablement is forwarded to provider stream options and re-resolved per continuation, keeping mid-run thinking-off changes in sync with the next provider request.

@oh-my-pi/pi-ai

Added

  • Added antigravityRankingStrategy and registered it for google-antigravity in DEFAULT_RANKING_STRATEGIES, so new sessions are routed to OAuth credentials with quota headroom for the requested model backend (lowest relevant remainingFraction counter as the sole ranked window, 24h windowDefaults matching daily-cloudcode-pa.googleapis.com resets). Without it, the existing antigravityUsageProvider data never reached credential selection. (#2198)

Changed

  • Updated MiniMax and MiniMax Token Plan defaults to MiniMax-M3 and refreshed Token Plan login copy/links (#1725).

Fixed

  • Fixed OpenAI Responses and Azure OpenAI Responses streams silently surfacing incomplete output as successful when a custom/proxy provider drops the connection without sending a terminal response.completed/response.incomplete event. Both providers now detect premature stream closure and throw with stopReason: "error" (#2184)
  • Fixed isUsageLimitError missing Antigravity / Cloud Code Assist's Individual quota reached 429 phrasing. The USAGE_LIMIT_PATTERN only knew quota.?exceeded / limit_reached, so auth-retry and AuthStorage.markUsageLimitReached treated the response as a terminal provider error and pinned sessions to the exhausted OAuth account instead of rotating to a sibling credential. The pattern now also matches quota.?reached. (#2198)
  • Scoped Antigravity usage blocking and ranking by model family (gemini-*/gemma-* → Google, claude-* → Anthropic, gpt-*/openai/* → OpenAI), so an exhausted Gemini counter no longer makes a healthy Claude/OpenAI Antigravity credential unavailable until reset. (#2198)
  • Fixed no-model Antigravity credential lookups (e.g. image-provider discovery) inheriting provider-wide exhaustion: scopeLimits now returns no limits without a concrete backend counter, and blockScope always returns a counter scope so missing model context can never fall through to AuthStorage's provider-wide block bucket. (#2198)

@oh-my-pi/pi-catalog

Added

  • Added grok-composer-2.5-fast (Cursor "Composer 2.5 Fast") to the xAI Grok OAuth (SuperGrok) catalog: non-reasoning, text-only, 200K context.

Changed

  • Set every xAI Grok OAuth (SuperGrok) curated model's max output tokens to mirror its context window (grok-build, grok-4.3, grok-4.20-0309-{reasoning,non-reasoning}, grok-4.20-multi-agent-0309, grok-composer-2.5-fast), replacing the 8888 UNK_MAX_TOKENS placeholder (and a stale 30000 on three grok-4.x entries). xAI's OAuth /v1/models reports no per-request output limit, so the curated catalog now owns maxTokens like contextWindow, deterministic on both the static-seed and online-overlay paths; the openai-responses wire still clamps the actual request to OPENAI_MAX_OUTPUT_TOKENS (64k).

Fixed

  • Excluded zero-cost xai-oauth subscription entries from the model reference indexes (buildModelReferenceIndex, createReferenceResolver), so their zero pricing and context-window-sized maxTokens cannot outrank paid/public Grok references when resolving custom-provider model identities.

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

Added

  • Added RPC subagent subscription frames, snapshots, and transcript catch-up APIs for desktop clients embedding omp --mode rpc.
  • Added opt-in shellMinimizer.sourceOutlineLevel and shellMinimizer.legacyFilters settings so shell minimization can tune source outlining and selectively fall back to conservative legacy routing.
  • Added repeatable --config <path> CLI overlays for temporary config.yml-style settings without editing the persistent global config (#1733).
  • Added python.interpreter to pin eval's Python backend to an explicit interpreter and skip automatic runtime discovery (#1802).
  • Added !command resolution for models.yml provider apiKey values and provider/model headers (#1888).
  • Documented the oMLX setup path through existing OpenAI-compatible local discovery (#1957).
  • Added TITLE_SYSTEM.md discovery so users can override the automatic session-title generation prompt for online and local tiny title models without patching installed prompt files. The override is re-discovered when the session working directory changes via /cwd.
  • Added a structured memory runtime surface for extensions and UI integrations to query backend status, search memories, and save explicit memories across the configured memory backend.
  • Added support for Git repositories using the reftable storage format by detecting extensions.refStorage = reftable in the repository configuration and falling back to shelling out to Git commands (git symbolic-ref, git rev-parse) for reference and HEAD resolution.
  • Added /setup providers (also available as /setup or /providers) to reopen the interactive provider setup scene from an active TUI session, letting users sign in and choose a web search provider without rerunning the full onboarding flow.

Changed

  • Bash execution now preserves minimized shell output inline while saving the untouched capture as an artifact://… footer when shell minimization rewrites a command's output.
  • Task tool live progress now renders finished subagents first and keeps unfinished (pending/running) ones pinned at the bottom of the list.
  • OutputSink artifact files (~/.omp/agent/artifacts/<id>.<tool>.log) are unbounded by default again, so artifact://<id> references preserve the complete raw stream. The head + rolling-tail capping machinery from #2081 (with its [ARTIFACT TRUNCATED: …] close notice) remains available as an opt-in via artifactMaxBytes, and the head window now closes permanently on first overflow so later small chunks cannot be written out of order before the tail replay.

Fixed

  • Fixed Ollama chat turns using the :off thinking selector so requests explicitly send reasoning disablement instead of falling back to the provider default (#2239).
  • Fixed long-running sessions becoming sluggish because the status line recomputed context usage by walking the full message history on every refresh. Message-token totals are now cached incrementally, and status lines that do not render context segments skip context accounting entirely. (#2089)
  • Fixed extension-registered slash commands being allowed to shadow builtin command names in the ACP/RPC command list: both the TUI and getSessionSlashCommands() now filter against the shared builtin reserved-name registry.
  • Fixed ACP turns for extension slash commands finishing before nested pi.sendUserMessage() prompts ran: the turn now drains scheduled extension prompts and prompt-event handlers before reporting end_turn, and prompts queued on a closed or disposed session fail fast with an ACP_SESSION_CLOSED error instead of running against a dead session.
  • Fixed hide-secrets placeholders leaking into ephemeral side-channel turns (IRC replies, summary prompts): streamed text deltas are now deobfuscated before emission (withheld while a partial placeholder is still streaming) and the final assistant message is deobfuscated before reuse. Provider-context obfuscation moved into the agent loop's transformProviderContext hook so request telemetry also captures the redacted context (#2146).
  • Fixed a failed Settings.init() permanently poisoning subsequent initialization: the cached init promise is now cleared on error so a retry can succeed.
  • Fixed exiting plan mode without confirmation only when neither the default plan file nor slug-named local plan files contain draft content (#2024).
  • Fixed enabledModels being ignored by the ACP model picker (Zed and other ACP clients) — AgentSession.getAvailableModels() now applies the configured allow-list, so only the models listed in enabledModels appear in the UI. Also applies consistently to the RPC get_available_models endpoint and the /model slash command. Glob selectors (anthropic/*), provider-scoped fuzzy patterns, and substring selectors now resolve in this synchronous path too, using the same scope semantics as startup resolution instead of being skipped with a warning.
  • ACP sessions now skip the client permission gate for bash/edit/delete/move when the user explicitly opts into yolo approval mode (--yolo/--auto-approve or a configured tools.approvalMode: yolo) and the effective per-tool policy is "allow"; default-config sessions keep the gate (#2097 by @Mokto)
  • Fixed hidden thinking blocks leaving placeholder Thinking... lines in the transcript (#2068).
  • Fixed Hindsight per-project-tagged scoping siloing retains/recalls per linked git worktree: projectLabel() now resolves the primary checkout root (or shared bare-repo common dir) via the new sync git.repo.primaryRootSync helper, so every worktree of one repo shares the same project:<name> tag and per-project bank id (#2232).
  • Fixed MCP OAuth flows accepting pasted redirect URLs or authorization codes through /login in headless environments (#2122).
  • Forwarded model ids through ModelRegistry API-key resolvers and Antigravity usage-limit rotation so pi-ai can apply model-family-scoped OAuth quota backoff instead of treating all google-antigravity counters as credential-wide. (#2198)
  • Fixed bare omp extensions being treated as a chat prompt instead of returning an actionable plugin-command error (#2089).
  • Fixed hide-secrets redaction so configured secrets are scrubbed from provider-facing system prompts, tool definitions, developer/system-reminder messages, and assistant tool-call arguments before model requests (#2146).
  • Fixed subagents looping indefinitely on byte-identical no-op edit calls. The hashline executor previously surfaced a soft "your body row(s) are byte-identical to the file" hint that some models ignored; one captured session emitted 182 such repeats in 205 calls over 16 minutes before the user aborted. A new per-ToolSession noopLoopGuard now tracks consecutive identical no-op payloads per canonical path and escalates to a thrown ToolError after NOOP_HARD_LIMIT (3) repeats, so the agent loop sees a tool failure and breaks the cycle (#2081).
  • Fixed the bash result renderer recomputing styled output (split / replaceTabs / truncateToVisualLines) on every TUI repaint, which scaled with both transcript length and per-row output size. With a long captured session every keystroke walked hundreds of bash rows; the reporter on issue #2081 observed Ctrl+X/Ctrl+C feeling unresponsive because the main thread was pinned re-styling scrollback. The result renderer now caches its produced lines keyed by (width, previewLines, expanded, rawOutput, isPartial), mirroring the existing eval-renderer cache; invalidate() clears the cache as before. Hot-path repaints with unchanged inputs are now O(1) (#2081).
  • Fixed ACP available_commands_update to include extension-registered slash commands so clients like Zed surface them in the slash-command palette.
  • Fixed ACP cancel button leaving the session in a stuck state — a new prompt sent while a turn is still in-flight (e.g. immediately after pressing Stop in Zed before session/cancel is processed) now implicitly cancels the running turn and queues the new message, instead of throwing an error that blocks further interaction.
  • Fixed interactive !/!! shell shortcuts to run non-bash commands through the configured user shell, including interactive startup for zsh/fish aliases and functions (#1816).

@oh-my-pi/pi-mnemopi

Changed

  • Reworked the in-memory fallback vector search to build a normalized exact vector index per query, matching the shape needed for future quantized or TurboVec-style backends without adding a new dependency yet.

@oh-my-pi/pi-natives

Added

  • Added deterministic shell-output minimization to the native shell pipeline, including opt-in per-command rewrite telemetry surfaced through executeShell().minimized for callers that want compact inline output plus a separately persisted original capture.

Fixed

  • Fixed native crash-log directory resolution diverging from the JS logger when PI_CONFIG_DIR is absolute: the config root now mirrors path.join(homedir, PI_CONFIG_DIR) semantics (absolute values re-rooted under $HOME, ./.. components normalized), and an empty PI_CODING_AGENT_DIR no longer disables XDG state-dir resolution.
  • Fixed shell-output minimization condensing pyright/basedpyright --outputjson runs into a diagnostics summary; machine-readable JSON output now passes through untouched.
  • Fixed pi-natives aborting Bun on Windows with memory allocation of N bytes failed and no backtrace whenever the native cdylib hit a Rust panic or out-of-memory condition. The release profile uses panic = "abort", so neither default handler emitted any context — Bun received only the bare message and tore down the TUI session before flushing. Module load now installs std::panic::set_hook and std::alloc::set_alloc_error_hook via #[napi::module_init]; both hooks capture Backtrace::force_capture() (so it works without RUST_BACKTRACE=1) and write a structured report — pid, thread, size/alignment for OOM, source location and message for panics, full backtrace — to the same logs directory the JS logger uses ($XDG_STATE_HOME/omp/logs/ on Linux/macOS when the user has migrated to XDG and PI_CODING_AGENT_DIR isn't customized, otherwise ~/.omp/logs/) and to stderr before the host process exits. The OOM hook prints the canonical allocation-failure line before any allocation-prone diagnostics and aborts immediately on re-entry, so real process-wide OOM still surfaces the fallback message instead of recursing in the report path (#2211).

What's Changed

  • fix(coding-agent): prefer daemonizing CLI clipboards over arboard on Linux by @roboomp in #2076
  • fix(coding-agent): cache status line context usage by @cexll in #2094
  • feat(memory): add runtime API and exact vector index by @DarkPhilosophy in #2108
  • fix(coding-agent): hide secrets in provider requests by @roboomp in #2147
  • fix(mcp): accept manual OAuth redirect input by @danzaio in #2160
  • fix(cli): reject ambiguous extensions command by @danzaio in #2161
  • docs(models): document oMLX OpenAI-compatible setup by @danzaio in #2162
  • fix(tui): omit hidden thinking placeholder by @danzaio in #2163
  • fix(plan): skip confirm for empty plan exit by @danzaio in #2164
  • feat(ai): add grok-composer-2.5-fast to the SuperGrok catalog by @metaphorics in #2173
  • feat(minimizer): deterministic shell-output minimizer by @GratefulDave in #2176
  • fix: prevent silent stream termination and spurious stall for custom response-compatible providers by @lanshi17 in #2184
  • fix(acp): auto-cancel in-flight turn when new prompt arrives mid-flight by @Mokto in #2186
  • fix(ai): rotate antigravity credentials on Individual quota reached 429s by @roboomp in #2200
  • feat: add title system prompt override by @zdearo in #2201
  • feat(config): add runtime config overlays by @danzaio in #2202
  • fix(ai): default MiniMax providers to M3 by @danzaio in #2203
  • feat(eval): add python interpreter setting by @danzaio in #2204
  • feat(models): resolve secrets from commands by @danzaio in #2205
  • fix(natives): captured panic and OOM diagnostics from cdylib by @roboomp in #2212
  • feat: add RPC subagent subscriptions by @korri123 in #2214
  • fix(hindsight): share project tag across git worktrees by @roboomp in #2233
  • feat(coding-agent): add support for git reftables by @AngerM in #2018
  • feat(coding-agent): add provider setup command by @handlecusion in #2019
  • fix: apply enabledModels filter to ACP model list (Zed model picker) by @Mokto in #2044
  • fix(acp): emit extension-registered commands in available_commands_update by @Mokto in #2052
  • fix(coding-agent): broke runaway edit loops and capped bash artifact spew by @roboomp in #2083
  • fix(acp): skip permission gate in yolo mode when effective policy is allow by @Mokto in #2097
  • fix(providers): disable Ollama thinking for off turns by @roboomp in #2240
  • fix(coding-agent): respect user shell for ! shortcuts by @handlecusion in #1896

New Contributors

Full Changelog: v15.10.11...v15.10.12

Check out latest releases or
releases around can1357/oh-my-pi v15.10.12

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

NewReleases is sending notifications on new releases.

Get notifications