- agent-loop: raise repetition-detection floor to 180 chars and clear thinking
replay anchors when collapsing a detected loop.
- providers/google: ignore empty text parts, retain terminal thoughtSignatures,
and stop function-call signatures clobbering the prior block.
- autolearn: capture goal-mode at the turn boundary; harden managed-skill writes
against hard-links/symlinks (O_NOFOLLOW + nlink); refuse minting managed skills
whose name an authored skill already claims.
- eager tasks: thread agentKind through the session so a custom top-level agentId
still gets always-mode delegation; split Eager Tasks prompt into hard vs soft.
- title-generator: race the online title model against a local tiny-model fallback.
- eager-todo: keep the soft reminder aligned with the todo init schema.
- mcp/stdio: keep close() detaching the read loop instead of awaiting it.
- stream loop: fix collapsing and tool-call thought-signature handling.
Redirected legacy pi-ai utils/oauth subpaths through the compatibility loader so background workers load the relocated OAuth registry modules.\n\nFixes #2566
The previous `anyBuiltInSkillSourceEnabled` mixed the new
`enableAgentsUser`/`enableAgentsProject` defaults with the named
third-party toggles, so a user who disabled Codex/Claude/Pi to silence
third-party CLI skill sources but kept the default `.agent[s]/skills`
toggles on still saw unknown providers (`opencode`, `github`,
`claude-plugins`, `gemini`) load via the fallback branch. The
`agents` provider already has its own explicit branch in
`isSourceEnabled`, so the fall-through gate now only inspects the named
third-party toggles. Adds a regression test that creates a fake
`~/.config/opencode/skills/leaked-opencode` and verifies it stays
filtered out when the third-party toggles are off and agents toggles
default to on.
Addresses PR #2405 review.
The managed-skill discovery block vetoes a name already in `skillMap` to let a
custom-directory authored skill win. Custom dirs never enter `result.all`, so
they are absent from `enabledAuthoredNames` — the map check is the only veto, and
it only works because the custom-dir loop runs first. Add a comment so a future
refactor doesn't invert the order.
Addresses review thread on PR #2542 (thread 7).
Add a default-off "auto-learn" loop. When `autolearn.enabled` is set, after the
agent stops a session controller nudges it to capture reusable lessons: durable
facts go to long-term memory and repeatable procedures become "managed skills" —
SKILL.md files written to an isolated ~/.omp/agent/managed-skills directory that is
discovered and surfaced like authored skills but never overwrites them.
Two tools back this:
- `manage_skill` — create/update/delete managed skills.
- `learn` — record a lesson, optionally minting/enhancing a managed skill in the
same call (requires a hindsight/mnemopi memory backend).
The nudge is passive by default (a hidden reminder rides the next turn);
`autolearn.autoContinue` instead auto-runs one capture turn at stop, and
`autolearn.minToolCalls` (default 5) gates trivial turns. Plan/goal-mode turns and
subagents are never nudged, and the controller re-checks the live setting at fire
time so a mid-session opt-out takes effect.
Isolation & precedence: managed skills are a separate lowest-priority discovery
provider, so an authored skill of the same name wins across every provider and
custom directory regardless of third-party toggles; a disabled higher-priority
authored skill can never hide a managed one, and managed never masks an enabled
authored skill. Managed names and descriptions are sanitized on both write and
read (control/format chars, angle brackets, and Markdown fences) before they render
into the system prompt, and the SKILL.md byte cap is enforced on the final
serialized file.
Default off → zero footprint when disabled.
- Added session-domain modules and exports for session-entries, context, listing, loader, and migrations.
- Changed persistence to async append writes plus writeTextAtomic, removing sync line APIs.
- Added compaction-aware session context rebuild with dangling tool-call cleanup.
- Added resumable session resolution with status inference, id/stem/suffix matching, and backup recovery.
- Switched extensibility loader imports from namespace-style `zod` imports to named `z` imports in `zod/v4`.
- Updated extensibility type interfaces to use `typeof z` for injected `zod` modules in hook, extension, tool, and command APIs.
Pre-fix `loadSkills` in `extensibility/skills.ts` gated the OMP-native
`agents` provider (the canonical `~/.agent[s]/skills` and project-walkup
`.agent[s]/skills`) on the `anyBuiltInSkillSourceEnabled` fall-through —
the OR of `enableCodexUser`/`enableClaudeUser`/`enableClaudeProject`/
`enablePiUser`/`enablePiProject`. A user who turned off the named
third-party toggles to clean noise from other CLIs silently lost their
own canonical OMP-native skills, so typing `/` showed no `/skill:<name>`
entries for `~/.agents/skills/*/SKILL.md`.
- Added `skills.enableAgentsUser` and `skills.enableAgentsProject` to
the settings schema (default true) and to the `SkillsSettings` shape.
- Routed `isSourceEnabled` to the new toggles for `provider === "agents"`
(per level). The fall-through still covers providers without a
dedicated toggle (`claude-plugins`, `opencode`, `gemini`, `github`, …);
agents joins claude/codex/native at the named-source layer where it
always belonged.
- Updated `docs/skills.md` to document the new toggles and the corrected
enablement rule.
- Refactored `test/skills.test.ts` to share a `DISABLE_ALL_BUILTIN_SKILLS`
helper (so the isolation tests stay correct when new toggles are
added) and added two regression tests: one asserting that the
pre-fix scenario (named third-party toggles off, agents defaulted) now
loads `~/.agents/skills/<name>/SKILL.md`, and one pinning the explicit
`enableAgentsUser: false` opt-out.
Fixes#2401
- Moved completion/hint builder functions and BUILTIN_SLASH_COMMANDS materialization into slash-commands/builtin-registry.ts.
- Removed those builder definitions from extensibility/slash-commands and shifted type exports to the types module.
- Updated interactive-mode to import BUILTIN_SLASH_COMMANDS from builtin-registry instead of the extensibility wrapper.
Validated npm plugin extension entry points before recording installs and rolled back fresh installs that cannot resolve their extension imports.
Fixes#2312
Adds compaction.strategy: "snapcompact" to the schema and the AgentSession routing: when chosen, both manual /compact (without custom instructions) and auto compaction call snapcompactCompact() to archive history as PNG frames instead of an LLM summary. Falls back to context-full with a visible warning notice when the current model is text-only or when /compact gets custom instructions. CustomTool and shared-event payloads carry the new action through. \n\nAlso wires the per-turn supersede pass: #pruneSupersededReads() runs every turn before threshold gating (cache-aware: only fires when the post-candidate suffix is small or the prompt cache is cold), prunes older read results superseded by a newer read of the same file, rewrites the session, and accounts the saved tokens in the next compaction decision. Gated by compaction.supersedeReads (default on).\n\nsession/messages.ts now delegates the core role conversion to agent-core's convertMessageToLlm so snapcompact image blocks flow through the LLM-context conversion path.
ACP_BUILTIN_SLASH_COMMANDS only carries primary names; the reserved set
passed to getRegisteredCommands was therefore missing aliases like
"models" (/model) and "force:" (/force). An extension registering one
of these aliases would appear in the palette but the builtin would win
at dispatch time (lookupBuiltinSlashCommand searches aliases too).
Export ACP_BUILTIN_RESERVED_NAMES from acp-builtins — the union of all
primary names and aliases for ACP-surfaced builtins — and use it as the
reserved set. Widen getRegisteredCommands parameter to ReadonlySet since
it only calls .has().
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Replaced minLevel/maxLevel range with explicit efforts array plus baked effortMap/supportsDisplay wire facts.
- Removed runtime enrichment layer and modelOmitsReasoningEffort; providers now read baked fields.
- Fixed dotted Opus 4.7/4.8 ids missing adaptive display via classifier-based predicates (#1373).
- Bumped model cache schema to v4 to invalidate pre-efforts rows.
- Centralized catalog and registry handling on `ModelSpec` and `buildModel`, resolving compatibility at model build time.
- Removed runtime compatibility detectors and switched provider request flows to direct `model.compat` reads.
- Added compat fields (`supportsReasoningParams`, `alwaysSendMaxTokens`, `strictResponsesPairing`, `whenThinking`).
- Persisted explicit compatibility overrides through `compatConfig` in discovery and cache merge paths.
- Rerouted sync, tab, js-eval, and tiny workers to re-enter CLI modes via `__omp_*` selectors.
- Adjusted `cli.ts` startup to dispatch worker entrypoints before parsing and exit 1 on uncaught errors.
- Bundled CLI as `dist/cli.js` in prepack, switching `omp` binary and published files.
- Removed explicit Bun `--compile` worker entrypoints from build/release scripts in favor of host-entry dispatch.
- Added `declareWorkerHostEntry()` and `workerHostEntry()` environment helpers and `PI_COMPILED` binary detection.
Same shape of bug the reviewer flagged for custom tools: forwarding
`LoadExtensionsResult` from parent to subagent reused Extension instances
whose factories closed over the parent's `ExtensionAPI` — cwd, eventBus,
and runtime all pointed at the parent. Any tool/handler/command that
referenced `api.exec()`, `api.events`, or `api.runtime` still acted on the
parent session/worktree from inside an isolated subagent.
Forward only the path list; each session rebuilds extensions through
`loadExtensions` so factories see the right `ExtensionAPI`.
- `extensibility/extensions/loader.ts`: extract `discoverExtensionPaths`
(FS scan only) from `discoverAndLoadExtensions`. The combined helper now
composes the two. New export added to the package barrel.
- `sdk.ts`:
- Add `discoverSessionExtensionPaths()` (the `disableExtensionDiscovery`-aware
path-only counterpart of `loadSessionExtensions`).
- Add `preloadedExtensionPaths?: string[]` to `CreateAgentSessionOptions`.
Three loader branches: `preloadedExtensions` (CLI same-process reuse,
still shallow-cloned), `preloadedExtensionPaths` (subagent: skip scan,
reload locally), or full discovery.
- Document `preloadedExtensions` as same-process-only; subagent
forwarding MUST use `preloadedExtensionPaths`.
- `tools/index.ts`: `ToolSession.extensionsResult` → `extensionPaths:
string[]` for the same reason.
- `task/executor.ts` and `task/index.ts`: forward `extensionPaths`. Drop
the forward for the isolated `runSubprocess` branch — worktree cwd ≠
parent cwd, so the subagent re-discovers extensions against its own
tree.
- New `test/sdk-extensions-per-session-binding.test.ts` pins the contract:
two `loadExtensions` calls on the same path with different `cwd` and
different `EventBus` instances yield distinct Extension + runtime
objects whose factories close over the per-call bindings.
- Updated `executor-pass-through` and `sdk-preloaded-extensions-isolation`
tests for the new option name and comment context.
Refs PR review on #2193
Reviewer flagged that forwarding `LoadedCustomTool[]` from a parent session
to a subagent reused tool instances whose factories had closed over the
parent's `CustomToolAPI` — `cwd`, `exec`, `pushPendingAction`, and `ui` all
pointed at the parent. In isolated tasks the tool would `exec` against the
parent worktree and queue pending actions on the parent session.
Forward only the path list; let each session rebuild tools through
`loadCustomTools` so factories see the right `CustomToolAPI`.
- `extensibility/custom-tools/loader.ts`: extract `discoverCustomToolPaths`
(FS scan only) from `discoverAndLoadCustomTools`; export
`ToolPathWithSource`. The combined helper is now `discoverCustomToolPaths`
+ `loadCustomTools`.
- `sdk.ts`: replace `preloadedCustomTools` (`LoadedCustomTool[]`) with
`preloadedCustomToolPaths` (`ToolPathWithSource[]`). The custom-tools
block runs `loadCustomTools` unconditionally; only the path scan is
skipped when the caller pre-discovered it.
- `tools/index.ts`: `ToolSession.loadedCustomTools` →
`ToolSession.customToolPaths` for the same reason.
- `task/executor.ts` and `task/index.ts`: forward `customToolPaths`.
Drop the forward for isolated subagents — the worktree shifts `cwd`, so
the subagent re-discovers tools against its own working tree.
- New `test/sdk-custom-tools-per-session-binding.test.ts` pins the contract:
two `loadCustomTools` calls on the same path with different `cwd` and
different `pushPendingAction` callbacks yield distinct tool instances
whose factories see the per-call bindings.
- Updated `executor-pass-through` and `sdk-preloaded-extensions-isolation`
tests for the new option name and added a `ToolPathWithSource` fixture.
Refs PR review on #2193
The package-root override branch of `resolveCanonicalPiSpecifier` returned
the bunfs override path without checking the target was actually present,
so when `bun --compile` quietly dropped one of the extra entrypoints
(observed on macOS arm64 release binaries), the static rewrite emitted a
`file://` URL to a missing module. The #1216 fallback only fired on the
throwing `getResolvedSpecifier` path, so the override never threw and the
rewrite committed the bad URL — extensions silently failed to load.
Each override target is now checked with `fs.existsSync` at module init.
Missing entries are dropped from `LEGACY_PI_PACKAGE_ROOT_OVERRIDES` so
`resolveCanonicalPiSpecifier` falls through to `getResolvedSpecifier`,
which throws under bunfs and triggers the existing rewrite catch — Bun
then resolves the canonical `@oh-my-pi/pi-*` specifier from the
extension's own `node_modules`.
Fixes#2168
- Added optional FetchImpl fields to compaction, proxy, AI, coding-agent, and mnemopi options.
- Threaded injected fetch implementations through OAuth, discovery, and search/LLM request flows.
- Removed exported hookFetch utility and its package entrypoint from utils.
- Replaced global-fetch test monkeypatching with per-test FetchImpl mocks across test suites.
- Derived descriptors, default-model map, env keys, login list, and refresh dispatch from one ProviderDefinition per provider.
- Disabled OpenAI Codex stream obfuscation and interrupted whitespace-only tool-call argument deltas.
- Derived auth-broker callback ports and paste-code login set from the registry.
- Added status.done and tool.* symbols to theme mappings and presets.
- Replaced generic success glyphs with contextual +/-, tool icons, and warnings.
- Mapped tool/task/job completions to status.done or status.enabled with icon overrides.
- Triggered runtime provider refresh after extension registration and warned on failure.
MCPTool and DeferredMCPTool now declare approval = 'write' instead of
implicitly defaulting to 'exec'. Without this, the approval system
requires user confirmation for every MCP tool call in non-yolo modes,
but the confirmation prompt never renders in the TUI while streaming,
causing the agent to hang indefinitely.
Also propagate the approval property through customToolToDefinition()
in sdk.ts, which was silently dropping it during CustomTool ->
ToolDefinition conversion.
- Deferred setup wizard import until setup was forced or version stale.
- Dynamically loaded ACP, RPC, and print mode runners only when used.
- Added a marketplace auto-update scheduler with off-mode early exit and non-blocking errors.
- Added setup-version assertions to keep CURRENT_SETUP_VERSION aligned with scenes.
- Added anonymous Perplexity authentication mode for unauthenticated web searches.
- Switched web-search setup checks to use `isExplicitlyAvailable` and removed key enforcement in doctor.
- Updated Perplexity OAuth flow to reuse auth handling for all non-key searches and anonymous responses.
- Updated CLI and provider option help text to mark the Perplexity key optional with fallback.
- Added `TUI.resetDisplay()` to force an immediate full-frame replay including native scrollback.
- Moved the persistent model selector default from Ctrl+L to Alt+M, preserving existing user remaps.
- Reserved Alt+M so extensions cannot shadow the model selector shortcut.
- Described "auto" default gating MCP tools past 40-tool threshold.
- Noted late resolution in createAgentSession after registry exists.
- Updated legacy mcp.discoveryMode mapping to MCP-only.
Kept explicit null package imports as exclusions so exact entries and active conditions do not fall through to wildcard or fallback targets.\n\nFixes #1889
ExtensionRunner#getShortcuts() accepted ctrl+q because #RESERVED_SHORTCUTS
predated the new default, and InputController registers extension
shortcuts before the followUp keybinding, so the editor's custom-key
map silently overwrote the extension handler. Now ctrl+q is reserved
alongside the other built-in chords and the extension authoring docs
list it as such.
Addresses code review on #1905.
Returned null from resolveSourceModuleFile when a package imports alias points at a JSON, WASM, or other non-code asset so the on-load rewrite hook no longer claims it and forces it through the JS loader. Bun resolves these targets natively. Added a regression where #schema maps to a .json file imported with a JSON type assertion.\n\nFixes #1889
Selected conditional package import targets by package.json object order for supported Bun runtime conditions, including node, instead of probing a fixed precedence list. Added a regression where node precedes import and must be selected for a plugin-local #src/* import.\n\nFixes #1889
Extended the legacy Pi, TypeBox, package-import alias, and extension graph regexes to recognise the bare \"import \\"specifier\\";\" shape so side-effect-only loads such as \"import \\"#src/register\\";\" walk into the source graph and get their legacy @(scope)/pi-* imports rewritten. Added a regression that loads a plugin with a side-effect alias import whose target contains a legacy scope import.\n\nFixes #1889
Resolved plugin-local package import aliases while loading extension source graphs so legacy Pi plugins with TypeScript source imports like #src/* register correctly under OMP. Added a regression covering legacy scope rewrites through a package import module.\n\nFixes #1889
can1357