Run next-turn preparation only when the agent loop continues, allowing threshold compaction before the next provider request. Restore interactive progress when the same run resumes.
Closes#6879
Important to notice:
this changes when prepareNextTurn & prepareNextTurnWithContext are invoked:
the now run only when a next turn actually happens, and after shouldStopAfterTurn.
Extension messages sent with triggerTurn: false while the agent was
running were pushed into agent state and the session tree immediately,
landing between an assistant tool call and its result. Providers that
validate message order reject the replayed history. Queue them during
the run and append at turn_end, mirroring recordBashResult.
closes#8537
* fix(coding-agent): keep model and thinking level changes session scoped
* fix(coding-agent): persist requested default thinking level
* feat: --default arg to persist
* fix
* fix(coding-agent): keep model and thinking level changes session scoped
* fix: make default thinking level per model show list of models
* cleanup export
Add provider-context prefix preparation and optional summary request settings needed by cache-friendly compaction. Existing AgentSession callers continue to pass undefined, so standalone summarization remains the default until the feature is wired in.
- add `--use-theme` for fixed and terminal-aware theme pairs
- keep invocation state non-persistent until an explicit selection
- preserve active themes across previews, reloads, and HTML exports
* fix(coding-agent): resume after context-limited length stops
Treat length stops as context overflow when prompt usage is within 1% of
Pi's configured context window. Include cache-write tokens in that usage
and allow non-zero output because OpenAI may emit reasoning tokens before
returning an incomplete response.
After compaction, remove any retryable error or truncated-length assistant
message restored from persisted session history so the interrupted agent
turn can continue.
This is specifically for handling this case for openai models:
https://developers.openai.com/api/docs/guides/reasoning
> If the generated tokens reach the context window limit or the max_output_tokens value you’ve set, you’ll receive a response with a status of incomplete and incomplete_details with reason set to max_output_tokens. This might occur before any visible output tokens are produced, meaning you could incur costs for input and reasoning tokens without receiving a visible response.
This detection currently assumes the reported prompt usage is within 1% of
the context window configured in Pi. Its reliability therefore depends on
Pi's model contextWindow metadata matching the provider's effective limit.
Maybe we should consider treating OpenAI incomplete responses as requiring compaction
regardless of the local context-window estimate?
fixes#7020
* fix(coding-agent): recover truncated responses after compaction
Classify length stops by comparing provider-reported output usage with the model's intended output limit. Compact and retry once when generation ends below that limit, avoiding unreliable context-window estimates while preserving genuine max-output stops.
Normalize OpenAI Responses incomplete reasons so only max_output_tokens is treated as a length stop. Preserve qualified provider reasons for diagnostics, surface other reasons as errors, and use neutral truncation wording in the TUI.
* fix(coding-agent): preserve agent events during compaction
Manual compaction already waits for the active agent run to settle, and summary generation does not emit Agent events. Keep the agent subscription connected so unexpected concurrent events are not dropped, and remove the obsolete reconnect helper.
closes#7253
Images produced by tools (extensions, MCP bridges, screenshot tools) went into
session history at full resolution because processImage was only wired to the
read tool and @file CLI attachments. Providers validate every image in the
request, so one oversized image poisons the whole conversation: Anthropic drops
its per-image cap from 8000px to 2000px once a request carries many images, and
every later turn re-sends the same payload. Switching models does not help.
Normalize tool result images in AgentSession.afterToolCall, after the extension
tool_result hook so extension-injected images are covered too. Honors
images.autoResize, and keeps the original block when processing fails so an
unavailable image backend does not delete tool output.
Refs #5369
Assisted-By: devx/86bf3b09-9aba-4d03-8154-50fca3ffc9b5
* fix(coding-agent): guard tree navigation during responses
* fix(coding-agent): simplify tree navigation regression
* fix(coding-agent): abort active run on session replacement and tree navigation
Session replacement (resume, new, fork, import) now settles the active
response in teardownCurrent before disposing the outgoing session, so the
aborted turn and its tool results persist instead of leaving a dangling
tool call. Tree navigation aborts the active response directly once the
user commits, without a confirmation gate.
Extensions have no way to read the session's scoped model set (the models
resolved from --models / enabledModels against the available catalogue --
the same list /scoped-models shows). The only model surfaces on
ExtensionContext are ctx.model (active) and ctx.modelRegistry.getAvailable()
(the whole catalogue, unfiltered), so an extension that wants a scoped
model picker must either enumerate everything or re-implement pi's scope
resolution (minimatch on provider/modelId / bare modelId + enabledModels +
--models) from disk -- duplicating core logic.
Add ctx.scopedModels: readonly ScopedModel[] to ExtensionContext, mirroring
how ctx.model is exposed (field -> bindCore -> lazy getter in createContext()).
Empty when no scoping is configured (all available models usable). Each entry
is { model, thinkingLevel? }.
- types.ts: ExtensionContext.scopedModels + ExtensionContextActions.getScopedModels
- runner.ts: field, bindCore wiring, lazy scopedModels getter
- agent-session.ts: provide getScopedModels: () => this._scopedModels
- test: assert ctx.scopedModels reflects the action (default [] + live ref)
- docs/extensions.md: document the new field
Co-authored-by: ngSoftware <alessandro@pungitore.ch>
fixes#6647
compaction (auto & manual) and branch summarization retry on transient failures.
use the same retry policy from settings.
emit events for the tui to show indication of retries
* add usage info to branch summary entries
* add usage to compaction entries
* allow custom tools to report llm usage in tool results
* allow observing and patching usage in tool_result hooks
* agent-harness: save usage in entries
for compaction, branch summaries and tool results
Move provider auth and OAuth flows onto pi-ai Models, compose models.json and extension overlays through ModelRuntime, and retain ModelRegistry as an extension compatibility facade.
Detect prompt cache misses per turn by comparing each assistant message's
cache reads against the previous request's prompt tokens (core/cache-stats.ts).
Significant misses emit a warning-colored transcript notice at the turn they
occur, noting idle gaps past the cache TTL and model switches when relevant.
/session gained cache statistics: a compact token/cache breakdown with hit
rate, a $-prefixed cost section with per-model cost breakdown, and the
cumulative cost re-billed due to cache misses.
The Message types require content to always be present, but untyped JS
extension tools, hand-built histories, and old or hand-edited session
files can violate that contract, crashing rendering, compaction, and
provider request conversion with 'content is not iterable'.
Normalize null/missing content to an empty array at the ingestion
boundaries instead of guarding every consumer:
- transformMessages (choke point before every provider request)
- createToolResultMessage in the agent loop
- session entry loading (message and custom_message entries)
- extension custom messages (sendCustomMessage, before_agent_start)
- message_end extension replacements
fixes#6259, fixes#6276
