deepagents-code 0.1.67__tar.gz → 0.1.69__tar.gz

{deepagents_code-0.1.67 → deepagents_code-0.1.69}/CHANGELOG.md

@@ -2,6 +2,42 @@
 
 # Deep Agents Code Changelog
 
+## [0.1.69](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.68...deepagents-code==0.1.69) (2026-09-14)
+
+### Features
+
+- Update `read_file` output formatting. ([#5648](https://github.com/langchain-ai/deepagents/pull/5648))
+- Surface DeepSeek V4.1 Flash in the model picker. ([#6254](https://github.com/langchain-ai/deepagents/pull/6254))
+- Surface locally tracked GitHub stacks in agent context. ([#6290](https://github.com/langchain-ai/deepagents/pull/6290))
+- Copy a model slug with Ctrl+click. ([#6243](https://github.com/langchain-ai/deepagents/pull/6243))
+- Show session length in the Debug Console. ([#6224](https://github.com/langchain-ai/deepagents/pull/6224))
+
+### Bug Fixes
+
+- Price nested usage with its own model and honor completions. ([#6251](https://github.com/langchain-ai/deepagents/pull/6251))
+- Drop stale Anthropic thinking blocks. ([#6300](https://github.com/langchain-ai/deepagents/pull/6300))
+- Isolate credentials used for user shell tracing. ([#6242](https://github.com/langchain-ai/deepagents/pull/6242))
+- Attribute dotenv configuration sources. ([#6222](https://github.com/langchain-ai/deepagents/pull/6222))
+- Expose unknown reasoning effort values. ([#6241](https://github.com/langchain-ai/deepagents/pull/6241))
+- Open the Debug Console at the bottom of the log. ([#6218](https://github.com/langchain-ai/deepagents/pull/6218))
+- Order Debug Console log filters. ([#6217](https://github.com/langchain-ai/deepagents/pull/6217))
+- Show the spinner during pre-stream turn setup. ([#6253](https://github.com/langchain-ai/deepagents/pull/6253))
+- Demote no-output hint suppression messages to debug logging. ([#6245](https://github.com/langchain-ai/deepagents/pull/6245))
+
+## [0.1.68](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.67...deepagents-code==0.1.68) (2026-09-10)
+
+### Features
+
+- Recommended GLM 5.3 models are now available in code workflows ([#6211](https://github.com/langchain-ai/deepagents/pull/6211)).
+
+### Bug Fixes
+
+- Added pricing support for Fireworks GLM-5.3 and GLM-5.3 Flash usage ([#6221](https://github.com/langchain-ai/deepagents/pull/6221), [#6223](https://github.com/langchain-ai/deepagents/pull/6223)).
+- Workspace switches now ask for confirmation before restarting the server ([#6178](https://github.com/langchain-ai/deepagents/pull/6178)).
+- Effort changes are now included in cache identity for OpenAI and Anthropic requests ([#6196](https://github.com/langchain-ai/deepagents/pull/6196)).
+- Standardized the recent-thread environment variable ([#6205](https://github.com/langchain-ai/deepagents/pull/6205)).
+- Silenced warnings when the themes config is absent ([#6220](https://github.com/langchain-ai/deepagents/pull/6220)).
+
 ## [0.1.67](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.66...deepagents-code==0.1.67) (2026-09-09)
 
 ### Features

{deepagents_code-0.1.67 → deepagents_code-0.1.69}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: deepagents-code
-Version: 0.1.67
+Version: 0.1.69
 Summary: Terminal coding agent built on the Deep Agents SDK - works with any model, with persistent memory, customizable skills, subagents, and approval controls.
 Project-URL: Homepage, https://www.langchain.com/dcode
 Project-URL: Documentation, https://docs.langchain.com/oss/deepagents/code/overview
@@ -30,21 +30,21 @@ Requires-Python: <4.0,>=3.12
 Requires-Dist: aiosqlite<1.0.0,>=0.22.1
 Requires-Dist: anyio<5.0.0,>=4.14.2
 Requires-Dist: deepagents-acp<1.0.0,>=0.0.10
-Requires-Dist: deepagents==0.7.13
+Requires-Dist: deepagents==0.7.14
 Requires-Dist: filelock<3.30,>=3.29.7
 Requires-Dist: genai-prices<0.2.0,>=0.1.6
 Requires-Dist: httpx<1.0.0,>=0.28.1
-Requires-Dist: langchain-anthropic<2.0.0,>=1.7.1
+Requires-Dist: langchain-anthropic<2.0.0,>=1.7.2
 Requires-Dist: langchain-google-genai<5.0.0,>=4.4.0
 Requires-Dist: langchain-mcp-adapters<1.0.0,>=0.3.2
-Requires-Dist: langchain-openai<2.0.0,>=1.6.1
+Requires-Dist: langchain-openai<2.0.0,>=1.6.2
 Requires-Dist: langchain-quickjs<0.4.0,>=0.3.4
 Requires-Dist: langchain<2.0.0,>=1.4.0
 Requires-Dist: langgraph-checkpoint-sqlite<4.0.0,>=3.1.1
 Requires-Dist: langgraph-cli[inmem]<1.0.0,>=0.4.31
 Requires-Dist: langgraph-runtime-inmem<1.0.0,>=0.34.0
 Requires-Dist: langgraph-sdk<1.0.0,>=0.4.4
-Requires-Dist: langsmith[sandbox]>=0.12.2
+Requires-Dist: langsmith[sandbox]>=0.12.4
 Requires-Dist: markdownify<2.0.0,>=1.2.3
 Requires-Dist: mcp<2.0.0,>=1.28.1
 Requires-Dist: packaging>=26.2
@@ -66,8 +66,8 @@ Provides-Extra: agentcore
 Requires-Dist: langchain-agentcore-codeinterpreter<1.0.0,>=0.0.5; extra == 'agentcore'
 Provides-Extra: all-providers
 Requires-Dist: aiohttp<3.15.0,>=3.14.3; extra == 'all-providers'
-Requires-Dist: langchain-anthropic<2.0.0,>=1.7.1; extra == 'all-providers'
-Requires-Dist: langchain-aws<2.0.0,>=1.7.5; extra == 'all-providers'
+Requires-Dist: langchain-anthropic<2.0.0,>=1.7.2; extra == 'all-providers'
+Requires-Dist: langchain-aws<2.0.0,>=1.7.6; extra == 'all-providers'
 Requires-Dist: langchain-baseten<1.0.0,>=0.2.4; extra == 'all-providers'
 Requires-Dist: langchain-cohere<1.0.0,>=0.6.0; extra == 'all-providers'
 Requires-Dist: langchain-deepseek<2.0.0,>=1.1.0; extra == 'all-providers'
@@ -82,7 +82,7 @@ Requires-Dist: langchain-meta<2.0.0,>=0.5.0; extra == 'all-providers'
 Requires-Dist: langchain-mistralai<2.0.0,>=1.1.6; extra == 'all-providers'
 Requires-Dist: langchain-nvidia-ai-endpoints<2.0.0,>=1.4.3; extra == 'all-providers'
 Requires-Dist: langchain-ollama<2.0.0,>=1.1.0; extra == 'all-providers'
-Requires-Dist: langchain-openai<2.0.0,>=1.6.1; extra == 'all-providers'
+Requires-Dist: langchain-openai<2.0.0,>=1.6.2; extra == 'all-providers'
 Requires-Dist: langchain-openrouter<2.0.0,>=0.2.8; extra == 'all-providers'
 Requires-Dist: langchain-perplexity<2.0.0,>=1.4.1; extra == 'all-providers'
 Requires-Dist: langchain-together<2.0.0,>=0.4.0; extra == 'all-providers'
@@ -94,11 +94,11 @@ Requires-Dist: langchain-modal>=0.0.5; extra == 'all-sandboxes'
 Requires-Dist: langchain-runloop>=0.0.6; extra == 'all-sandboxes'
 Requires-Dist: langchain-vercel-sandbox>=0.0.1; extra == 'all-sandboxes'
 Provides-Extra: anthropic
-Requires-Dist: langchain-anthropic<2.0.0,>=1.7.1; extra == 'anthropic'
+Requires-Dist: langchain-anthropic<2.0.0,>=1.7.2; extra == 'anthropic'
 Provides-Extra: baseten
 Requires-Dist: langchain-baseten<1.0.0,>=0.2.4; extra == 'baseten'
 Provides-Extra: bedrock
-Requires-Dist: langchain-aws<2.0.0,>=1.7.5; extra == 'bedrock'
+Requires-Dist: langchain-aws<2.0.0,>=1.7.6; extra == 'bedrock'
 Provides-Extra: cohere
 Requires-Dist: langchain-cohere<1.0.0,>=0.6.0; extra == 'cohere'
 Provides-Extra: daytona
@@ -132,7 +132,7 @@ Requires-Dist: langchain-nvidia-ai-endpoints<2.0.0,>=1.4.3; extra == 'nvidia'
 Provides-Extra: ollama
 Requires-Dist: langchain-ollama<2.0.0,>=1.1.0; extra == 'ollama'
 Provides-Extra: openai
-Requires-Dist: langchain-openai<2.0.0,>=1.6.1; extra == 'openai'
+Requires-Dist: langchain-openai<2.0.0,>=1.6.2; extra == 'openai'
 Provides-Extra: openrouter
 Requires-Dist: langchain-openrouter<2.0.0,>=0.2.8; extra == 'openrouter'
 Provides-Extra: perplexity

{deepagents_code-0.1.67 → deepagents_code-0.1.69}/deepagents_code/_build_info.py

@@ -1,3 +1,3 @@
 """Generated at build time. Do not edit or commit."""
 
-BUILD_COMMIT = "d0fbc7c"
+BUILD_COMMIT = "1d3232c"

{deepagents_code-0.1.67 → deepagents_code-0.1.69}/deepagents_code/_env_vars.py

@@ -506,6 +506,9 @@ hostile values the dotenv denylist does not yet enumerate. The global
 repo file, so a project `.env` cannot disable itself.
 """
 
+RECENT_THREADS = "DEEPAGENTS_CODE_RECENT_THREADS"
+"""Maximum number of recent threads loaded and displayed (default: 20)."""
+
 RECURSION_LIMIT = "DEEPAGENTS_CODE_RECURSION_LIMIT"
 """Override the main agent's LangGraph `recursion_limit` (graph step budget).
 

{deepagents_code-0.1.67 → deepagents_code-0.1.69}/deepagents_code/_session_stats.py

@@ -149,6 +149,32 @@ class RecordedUsage:
     request_tokens: int
     """Running token total for the request after applying this message."""
 
+    request_id: str | None = None
+    """Opaque key naming the request this delta belongs to, when known.
+
+    Keys each correction to the request it applies to. A late revision then
+    cannot subtract spend that belongs to other requests. `None` when the
+    message carried no usable ID, which is also when the ledger refuses to key
+    the request at all.
+
+    Built by `_provisional_bucket_key`, so it separates retry attempts exactly
+    as the ledger does. Treat it as a bucket key, not as a provider message ID.
+    """
+
+    is_correction: bool = False
+    """Whether this delta only revises spend already reported for the request.
+
+    A correction re-states what an earlier delta already contributed: a
+    completion replacing its chunks, or a request re-priced under the model it
+    finally named. New spend is not a correction, even when it arrives on a
+    request that already has a record, because the tokens themselves are new.
+
+    Lets the consumer drop a correction whose subject is gone -- a backend
+    total has already absorbed that request -- whatever the correction's sign.
+    A stale *positive* correction would otherwise re-inflate a display the
+    total had just settled.
+    """
+
 
 UsageLedgerKey = str | tuple[Hashable, str]
 """Key of the recorded-request ledger: a message ID, optionally scoped.
@@ -460,6 +486,36 @@ def finalize_recorded_requests(
             recorded_requests[request_id[1]] = closed
 
 
+def _provisional_bucket_key(request_id: UsageLedgerKey | None) -> str | None:
+    """Return a string key naming the request a provisional delta belongs to.
+
+    The consumer keys a pool of provisional dollars by this value, so it must
+    separate exactly what the ledger separates. An attempt-scoped key therefore
+    keeps its scope: two attempts that reuse one provider message ID are
+    distinct requests, and collapsing them to the bare ID would let one
+    attempt's late retraction draw on the other attempt's deposit.
+
+    Scope cannot change under a request while it still reports deltas. A chunk
+    or completion whose record is already finalized is rejected before any
+    delta is built, and `finalize_recorded_requests` closes every entry at the
+    end of a stream round -- so one request's deltas all carry one scope, and
+    the key stays stable for as long as the pool holds its money.
+
+    Args:
+        request_id: Ledger key for the request, or `None` when the message
+            carried no usable ID.
+
+    Returns:
+        An opaque bucket key, or `None` when there is no ledger key to name.
+    """
+    if request_id is None:
+        return None
+    if isinstance(request_id, tuple):
+        attempt_scope, message_id = request_id
+        return f"{attempt_scope!r}\x00{message_id}"
+    return request_id
+
+
 def _names_a_model(message: object) -> bool:
     """Report whether a message's own metadata names the model that served it.
 
@@ -630,21 +686,24 @@ def _resolve_usage_model(
         The `(model_name, provider)` pair to record and price under.
     """
     from deepagents_code.cost_tracking import (
+        _CONFIGURED_MODEL_METADATA_KEY,
         _CONFIGURED_PROVIDER_METADATA_KEY,
         resolve_message_model,
     )
 
-    configured_provider = (
-        request_metadata.get(_CONFIGURED_PROVIDER_METADATA_KEY)
-        if request_metadata is not None
-        else None
-    )
+    request_metadata = request_metadata or {}
+    configured_provider = request_metadata.get(_CONFIGURED_PROVIDER_METADATA_KEY)
     has_request_provider = isinstance(configured_provider, str) and bool(
         configured_provider
     )
+    configured_model = request_metadata.get(_CONFIGURED_MODEL_METADATA_KEY)
+    has_request_model = isinstance(configured_model, str) and bool(configured_model)
     return resolve_message_model(
         message,
-        fallback_model=fallback_model,
+        # The request's own configured model describes this very call, so unlike
+        # the caller's fallback it is valid for nested calls too; it only
+        # replaces the parent-fallback step, ahead of the existing chain.
+        fallback_model=configured_model if has_request_model else fallback_model,
         fallback_provider=(
             configured_provider if has_request_provider else fallback_provider
         ),
@@ -733,6 +792,117 @@ def _move_request_to_named_model(
     )
 
 
+def _finalize_from_completed(
+    stats: SessionStats,
+    message: object,
+    previous: RecordedRequest,
+    *,
+    usage: Mapping[str, Any],
+    recorded_requests: dict[UsageLedgerKey, RecordedRequest],
+    request_id: UsageLedgerKey,
+    fallback_model: str,
+    fallback_provider: str,
+    request_metadata: Mapping[str, Any] | None,
+    kind: UsageKind,
+) -> RecordedUsage | None:
+    """Replace a chunk-built request record with a completed message's usage.
+
+    Chunks report incremental deltas, so a request recorded mid-stream holds
+    partial tokens priced under whatever model was known then. The completed
+    message carries the whole usage; it must replace the partial record rather
+    than join it, or corrected token counts, cache details, and model
+    attribution never reach the stats.
+
+    Args:
+        stats: Accumulator holding the partial request.
+        message: The completed message with the request's whole usage.
+        previous: The partial ledger entry to supersede.
+        usage: The completed message's whole usage metadata.
+        recorded_requests: Ledger to update in place.
+        request_id: Message ID keying the ledger entry.
+        fallback_model: Model to use when response metadata does not name one.
+        fallback_provider: Provider to use when response metadata omits it.
+        request_metadata: Stream metadata for this specific request, if any.
+        kind: Request class used by the type breakdown.
+
+    Returns:
+        The signed cost change the replacement applied, for the caller's
+            provisional display, or `None` when nothing changed.
+    """
+    from deepagents_code.cost_tracking import cache_token_counts, estimate_cost
+
+    model_name, provider = _resolve_usage_model(
+        message,
+        fallback_model=fallback_model,
+        fallback_provider=fallback_provider,
+        request_metadata=request_metadata,
+        kind=kind,
+    )
+    if not _names_a_model(message):
+        # An earlier chunk may already have upgraded the record off the
+        # caller's fallback. A completion that names no model of its own must
+        # not undo that, or the request goes back to the parent's pricing --
+        # the very mis-attribution the chunk path guards against.
+        model_name, provider = previous.model_name, previous.provider
+    input_count, output_count = _display_token_counts(usage)
+    cost_usd = estimate_cost(usage, model_name, provider)
+    cache_reads, cache_writes = cache_token_counts(usage)
+
+    if (
+        not input_count
+        and not output_count
+        and _carries_token_counts(previous.usage_metadata)
+    ):
+        # A completion that corrects its chunks downward is ordinary -- the
+        # provider revises an over-counted prompt. Reporting *no* token field
+        # at all is not: it normalizes to zero and wipes a request the chunks
+        # had counted. Replace the record regardless, since the completion is
+        # authoritative, but say so rather than losing the tokens quietly.
+        logger.warning(
+            "A completed message states no token counts, so it replaces its "
+            "chunks with zero. request_id=%r dropped=%d/%d",
+            request_id,
+            previous.input_tokens,
+            previous.output_tokens,
+        )
+    stats.retract_request(previous)
+    stats.record_request(
+        model_name,
+        input_count,
+        output_count,
+        provider,
+        cost_usd=cost_usd,
+        kind=kind,
+        cache_read_tokens=cache_reads,
+        cache_write_tokens=sum(cache_writes),
+    )
+    recorded_requests[request_id] = RecordedRequest(
+        model_name=model_name,
+        provider=provider,
+        kind=kind,
+        input_tokens=input_count,
+        output_tokens=output_count,
+        cache_read_tokens=cache_reads,
+        cache_write_tokens=sum(cache_writes),
+        cost_usd=cost_usd,
+        usage_metadata=usage,
+        finalized=True,
+    )
+    return RecordedUsage(
+        input_tokens=input_count - previous.input_tokens,
+        output_tokens=output_count - previous.output_tokens,
+        cost_usd=_cost_delta(
+            previous.cost_usd,
+            cost_usd,
+            model_name=model_name,
+            previous_model_name=previous.model_name,
+        ),
+        request_tokens=input_count + output_count,
+        request_id=_provisional_bucket_key(request_id),
+        is_correction=True,
+    )
+
+
 def record_message_usage(
     stats: SessionStats,
     message: object,
@@ -810,10 +980,28 @@ def record_message_usage(
         request_id = (attempt_scope, request_id)
     is_chunk = isinstance(message, AIMessageChunk)
     if request_id is not None and request_id in recorded_requests and not is_chunk:
-        # A completed message repeats the whole request. Whether the request was
-        # built from chunks or from an identical earlier replay, it is already
-        # accounted for.
-        return None
+        previous = recorded_requests[request_id]
+        if previous.finalized:
+            # A completed message repeats the whole request, and a completed
+            # record already holds it in full. Whether this entry was built
+            # from an identical earlier completion or re-records one, it is
+            # already accounted for.
+            return None
+        # A partial ledger entry was built only from incremental chunks; this
+        # completion carries the request's whole usage and must *replace* it.
+        # Retry-attempt scoping keeps this keyed to its own attempt's entry.
+        return _finalize_from_completed(
+            stats,
+            message,
+            previous,
+            usage=usage,
+            recorded_requests=recorded_requests,
+            request_id=request_id,
+            fallback_model=fallback_model,
+            fallback_provider=fallback_provider,
+            request_metadata=request_metadata,
+            kind=kind,
+        )
 
     input_count, output_count = _display_token_counts(usage)
     previous = recorded_requests.get(request_id) if request_id is not None else None
@@ -850,6 +1038,8 @@ def record_message_usage(
                     output_tokens=0,
                     cost_usd=reprice_delta,
                     request_tokens=previous.input_tokens + previous.output_tokens,
+                    request_id=_provisional_bucket_key(request_id),
+                    is_correction=True,
                 )
         return None
 
@@ -925,6 +1115,7 @@ def record_message_usage(
             previous_model_name=previous.model_name if previous else model_name,
         ),
         request_tokens=input_count + output_count,
+        request_id=_provisional_bucket_key(request_id),
     )
 
 

{deepagents_code-0.1.67 → deepagents_code-0.1.69}/deepagents_code/_version.py

@@ -2,7 +2,7 @@
 
 # Keep the `x-release-please-version` annotation — release-please uses it to
 # bump `__version__` in sync with `pyproject.toml` on every release PR.
-__version__ = "0.1.67"  # x-release-please-version
+__version__ = "0.1.69"  # x-release-please-version
 
 DOCS_URL = "https://docs.langchain.com/oss/python/deepagents/code"
 """URL for `deepagents-code` documentation."""