Skip to content

fix(ai): record tool outcome on Claude Agent SDK and LangChain tool spans - #1282

Open
DanielVisca wants to merge 2 commits into
mainfrom
perch/ai-tool-span-outcomes
Open

DanielVisca wants to merge 2 commits into
mainfrom
perch/ai-tool-span-outcomes

Conversation

@DanielVisca

@DanielVisca DanielVisca commented Nov 20, 2026 •

Copy link
Copy Markdown
Contributor

Problem

Tool spans from two AI integrations can't tell you whether the tool worked, so per-tool failure rates can't be computed:

  • Claude Agent SDK (posthog.ai.claude_agent_sdk): the tool $ai_span was captured as soon as the ToolUseBlock was seen, before the tool ran. It never had $ai_output_state, $ai_is_error, $ai_error or $ai_latency. This affected both query() / PostHogClaudeAgentProcessor and PostHogClaudeSDKClient.
  • LangChain (CallbackHandler): tool runs were captured as untyped $ai_spans (no $ai_span_type: "tool"). A tool error handled by handle_tool_error ends the run normally with a ToolMessage(status="error"), so it was recorded as a success. ToolMessages in $ai_input also lost tool_call_id, name and status, so tool results couldn't be matched to tool names.

Changes

Claude Agent SDK (processor.py, client.py)

  • A tool span is now held by tool_use_id and emitted when its matching ToolResultBlock arrives. It then carries:
    • $ai_output_state: the result content, redacted and truncated like tool results in $ai_input
    • $ai_is_error: from block.is_error
    • $ai_error: the result text, only when the tool errored
    • $ai_latency: time from tool use to tool result
  • Privacy mode: $ai_is_error and $ai_latency are still sent; the output and error text are not.
  • Results inside subagent messages also close their spans. Previously subagent UserMessages were skipped entirely, but their tool spans were still emitted.
  • Spans whose result never arrives are still emitted, without an outcome:
    • query(): on ResultMessage, or in a finally if the stream ends early, fails or is cancelled
    • PostHogClaudeSDKClient: on ResultMessage and on disconnect()
  • The query() loop is wrapped in try/finally, which re-indents it. Review with whitespace changes hidden.

LangChain (callbacks.py)

  • Tool runs set $ai_span_type: "tool". Chains, retrievers and agent steps are unchanged.
  • In on_tool_end, a ToolMessage output with status == "error" sets $ai_is_error: True and $ai_error to the message text. $ai_error is None in privacy mode.
  • _convert_message_to_dict keeps tool_call_id and name on tool messages, plus status when it is "error".

Also adds a sampo changeset (patch).

Before / After

Printed by a small script that runs each integration against a MagicMock PostHog client and prints the $ai_span properties: (a) a Claude Agent SDK Bash tool call whose ToolResultBlock has is_error=True; (b) a LangChain @tool that raises ToolException with handle_tool_error=True, invoked with a tool call inside a chain. Trimmed to the relevant keys.

Before

(a) Claude Agent SDK $ai_span:
[{"$ai_span_name": "Bash", "$ai_span_type": "tool", "$ai_input_state": {"command": "foo"}}]
(b) LangChain $ai_span:
[{"$ai_span_name": "get_weather",
  "$ai_input_state": "{'city': 'Paris'}",
  "$ai_output_state": "content='weather service unavailable for Paris' name='get_weather' tool_call_id='call_1' status='error'",
  "$ai_latency": 0.0003371238708496094}]

After

(a) Claude Agent SDK $ai_span:
[{"$ai_span_name": "Bash", "$ai_span_type": "tool", "$ai_input_state": {"command": "foo"},
  "$ai_output_state": "bash: foo: command not found",
  "$ai_is_error": true,
  "$ai_error": "bash: foo: command not found",
  "$ai_latency": 0.003941774368286133}]
(b) LangChain $ai_span:
[{"$ai_span_name": "get_weather", "$ai_span_type": "tool",
  "$ai_input_state": "{'city': 'Paris'}",
  "$ai_output_state": "content='weather service unavailable for Paris' name='get_weather' tool_call_id='call_1' status='error'",
  "$ai_is_error": true,
  "$ai_error": "weather service unavailable for Paris",
  "$ai_latency": 0.0005731582641601562}]

Tests

  • posthog/test/ai/claude_agent_sdk/test_processor.py:
    • a successful result and an errored result (list content)
    • privacy mode
    • a subagent's tool result
    • a span flushed when the query ends without a result
    • spans emitted in order (generation → span → generation → trace)
  • posthog/test/ai/claude_agent_sdk/test_client.py: tool results arriving out of order match their own spans, and a span without a result is still emitted. Both run against query() and PostHogClaudeSDKClient.
  • posthog/test/ai/langchain/test_callbacks.py:
    • $ai_span_type is set on tool spans only
    • a handled tool error, with and without privacy mode
    • an unhandled tool error
    • ToolMessage conversion

uv run pytest posthog/test/ai/claude_agent_sdk posthog/test/ai/langchain: 251 passed, 8 skipped. ruff format --check, ruff check and mypy | mypy-baseline filter are clean.

Open questions for reviewers

  • $ai_output_state on Claude tool spans uses the same 5000-character truncation as tool results in $ai_input.
  • LangChain $ai_output_state for a ToolMessage is still its str() form (unchanged). I left it alone to keep this PR small.
  • Claude tool spans are now emitted when the result arrives rather than when the tool use is seen. Their timestamps move later, and they now arrive after the generation that called the tool.

🤖 Generated with Claude Code

…pans

Claude Agent SDK: hold each tool $ai_span until its ToolResultBlock arrives
and emit it with $ai_output_state, $ai_is_error, $ai_error and $ai_latency.
Spans without a result are flushed on ResultMessage, when the query stream
ends, and on client disconnect. Applies to query() and PostHogClaudeSDKClient.

LangChain: set $ai_span_type "tool" on tool spans, mark tool errors handled
by handle_tool_error (ToolMessage status "error") with $ai_is_error and
$ai_error, and keep tool_call_id, name and error status on ToolMessages.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@DanielVisca
DanielVisca requested a review from a team as a code owner November 20, 2026 03:05
@veria-ai

veria-ai Bot commented Nov 20, 2026 •

Copy link
Copy Markdown

PR overview

No security issues were found in this pull request.

Reviews

@posthog posthog Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

PostHog Review · Standard · reviewed 4d24e9d: 3 Should fix.

See this review and its findings in PostHog Review. Comment @posthog review --deep for a Deep review. A Deep review takes longer and looks deeper than a Standard review.

Comment thread posthog/ai/langchain/callbacks.py Resolved
Comment thread posthog/ai/langchain/callbacks.py Resolved
@github-actions

github-actions Bot commented Nov 20, 2026 •

Copy link
Copy Markdown
Contributor

SDK performance: posthog-python

Benchmarks completed. No changes flagged for review.

Flags require at least a 5% change in the median and all six pairs agreeing on direction. This is a noise filter, not proof of a performance change or a CI gate.

View all results
Test Measurement Before PR With PR Change Assessment
Capture and deliver events (32-byte payload, burst)
Shard: capture
Managed heap after settling — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​; head:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Managed heap after settling — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​; head:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​
Capture and deliver events (32-byte payload, steady)
Shard: capture
Managed heap after settling — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​; head:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Managed heap after settling — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​; head:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​
Queue an event (32-byte payload)
Shard: capture
Allocated memory per operation — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ Background preparation has no public completion boundary before flush; see capture.​drain for full-pipeline allocations.​; head:​ Background preparation has no public completion boundary before flush; see capture.​drain for full-pipeline allocations.​
Queue an event (32-byte payload)
Shard: capture
Allocations per operation — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ Background preparation has no public completion boundary before flush; see capture.​drain for full-pipeline allocations.​; head:​ Background preparation has no public completion boundary before flush; see capture.​drain for full-pipeline allocations.​
Queue an event (4,096-byte payload)
Shard: capture
Allocated memory per operation — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ Background preparation has no public completion boundary before flush; see capture.​drain for full-pipeline allocations.​; head:​ Background preparation has no public completion boundary before flush; see capture.​drain for full-pipeline allocations.​
Queue an event (4,096-byte payload)
Shard: capture
Allocations per operation — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ Background preparation has no public completion boundary before flush; see capture.​drain for full-pipeline allocations.​; head:​ Background preparation has no public completion boundary before flush; see capture.​drain for full-pipeline allocations.​
Evaluate 1,000 flags under concurrent load
Shard: flags-memory
Managed heap after settling — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​; head:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​
Refresh 1,000 flag definitions
Shard: flags-memory
Managed heap after settling — — — unavailable; base:​ unsupported; head:​ unsupported; base:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​; head:​ CPython has no uninstrumented total managed-heap byte counter; tracing would distort this RSS pass.​
Capture and deliver events (32-byte payload, burst)
Shard: capture
Allocated memory per operation 133.67 KiB/op 133.67 KiB/op No observed change Not flagged
Capture and deliver events (32-byte payload, burst)
Shard: capture
Allocations per operation 1,345.53 count/op 1,345.53 count/op No observed change Not flagged
Capture and deliver events (32-byte payload, burst)
Shard: capture
Time per operation 136.38 µs/op 143.57 µs/op 5.3% more time Not flagged
Capture and deliver events (32-byte payload, burst)
Shard: capture
Idle process memory (RSS) 37.77 MiB 37.66 MiB 0.3% less memory Not flagged
Capture and deliver events (32-byte payload, burst)
Shard: capture
Observed peak process memory (RSS) 40.21 MiB 40.11 MiB 0.3% less memory Not flagged
Capture and deliver events (32-byte payload, burst)
Shard: capture
Process memory after settling (RSS) 40.21 MiB 40.11 MiB 0.3% less memory Not flagged
Capture and deliver events (32-byte payload, burst)
Shard: capture
Process CPU time per operation 136.09 µs/op 145.35 µs/op 6.8% more time Not flagged
Capture and deliver events (32-byte payload, burst)
Shard: capture
Lifetime peak process memory (RSS) 40.45 MiB 40.45 MiB No observed change Not flagged
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Allocated memory per operation 185.76 KiB/op 185.75 KiB/op <0.1% less memory Not flagged
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Allocations per operation 1,337.56 count/op 1,337.53 count/op <0.1% fewer Not flagged
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Time per operation 169.54 µs/op 171.77 µs/op 1.3% more time Not flagged
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Idle process memory (RSS) 41.38 MiB 41.36 MiB <0.1% less memory Not flagged
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Observed peak process memory (RSS) 47.16 MiB 47.14 MiB <0.1% less memory Not flagged
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Process memory after settling (RSS) 47.16 MiB 47.14 MiB <0.1% less memory Not flagged
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Process CPU time per operation 160.95 µs/op 163.63 µs/op 1.7% more time Not flagged
Capture and deliver events (4,096-byte payload, burst)
Shard: capture
Lifetime peak process memory (RSS) 51.6 MiB 51.52 MiB 0.2% less memory Not flagged
Capture and deliver events (32-byte payload, steady)
Shard: capture
Allocated memory per operation 138.74 KiB/op 138.74 KiB/op No observed change Not flagged
Capture and deliver events (32-byte payload, steady)
Shard: capture
Allocations per operation 1,390.44 count/op 1,390.44 count/op No observed change Not flagged
Capture and deliver events (32-byte payload, steady)
Shard: capture
Time per operation 165.57 µs/op 153.7 µs/op 7.2% less time Not flagged
Capture and deliver events (32-byte payload, steady)
Shard: capture
Idle process memory (RSS) 37.59 MiB 37.69 MiB 0.3% more memory Not flagged
Capture and deliver events (32-byte payload, steady)
Shard: capture
Observed peak process memory (RSS) 37.77 MiB 37.85 MiB 0.2% more memory Not flagged
Capture and deliver events (32-byte payload, steady)
Shard: capture
Process memory after settling (RSS) 37.77 MiB 37.85 MiB 0.2% more memory Not flagged
Capture and deliver events (32-byte payload, steady)
Shard: capture
Process CPU time per operation 166.74 µs/op 151.87 µs/op 8.9% less time Not flagged
Capture and deliver events (32-byte payload, steady)
Shard: capture
Lifetime peak process memory (RSS) 40.45 MiB 40.45 MiB No observed change Not flagged
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Allocated memory per operation 190.92 KiB/op 190.92 KiB/op No observed change Not flagged
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Allocations per operation 1,382.36 count/op 1,382.36 count/op No observed change Not flagged
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Time per operation 175.42 µs/op 174.48 µs/op 0.5% less time Not flagged
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Idle process memory (RSS) 41.15 MiB 41.2 MiB 0.1% more memory Not flagged
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Observed peak process memory (RSS) 41.6 MiB 41.65 MiB 0.1% more memory Not flagged
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Process memory after settling (RSS) 41.21 MiB 41.25 MiB <0.1% more memory Not flagged
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Process CPU time per operation 166.68 µs/op 165.99 µs/op 0.4% less time Not flagged
Capture and deliver events (4,096-byte payload, steady)
Shard: capture
Lifetime peak process memory (RSS) 41.57 MiB 41.56 MiB <0.1% less memory Not flagged
Queue an event (32-byte payload)
Shard: capture
Caller-thread CPU time per operation 112.59 µs/op 109.65 µs/op 2.6% less time Not flagged
Queue an event (32-byte payload)
Shard: capture
Time per operation 113.4 µs/op 113.68 µs/op 0.2% more time Not flagged
Queue an event (4,096-byte payload)
Shard: capture
Caller-thread CPU time per operation 114.14 µs/op 109.05 µs/op 4.5% less time Not flagged
Queue an event (4,096-byte payload)
Shard: capture
Time per operation 121.46 µs/op 112.72 µs/op 7.2% less time Not flagged
Evaluate all 10 flags
Shard: flags-scaling
Allocated memory per operation 225.43 KiB/op 225.43 KiB/op No observed change Not flagged
Evaluate all 10 flags
Shard: flags-scaling
Allocations per operation 2,279.42 count/op 2,279.42 count/op No observed change Not flagged
Evaluate all 10 flags
Shard: flags-scaling
Time per operation 82.4 µs/op 82.37 µs/op <0.1% less time Not flagged
Evaluate all 10 flags
Shard: flags-scaling
Process CPU time per operation 82.4 µs/op 82.37 µs/op <0.1% less time Not flagged
Evaluate all 100 flags
Shard: flags-scaling
Allocated memory per operation 2.17 MiB/op 2.17 MiB/op No observed change Not flagged
Evaluate all 100 flags
Shard: flags-scaling
Allocations per operation 22,413.69 count/op 22,413.69 count/op No observed change Not flagged
Evaluate all 100 flags
Shard: flags-scaling
Time per operation 772.03 µs/op 770.9 µs/op 0.1% less time Not flagged
Evaluate all 100 flags
Shard: flags-scaling
Process CPU time per operation 771.77 µs/op 770.94 µs/op 0.1% less time Not flagged
Evaluate all 1,000 flags
Shard: flags-scaling
Allocated memory per operation 21.63 MiB/op 21.63 MiB/op No observed change Not flagged
Evaluate all 1,000 flags
Shard: flags-scaling
Allocations per operation 223,655.44 count/op 223,655.44 count/op No observed change Not flagged
Evaluate all 1,000 flags
Shard: flags-scaling
Time per operation 7.78 ms/op 7.7 ms/op 1% less time Not flagged
Evaluate all 1,000 flags
Shard: flags-scaling
Process CPU time per operation 7.78 ms/op 7.7 ms/op 1% less time Not flagged
Evaluate 1,000 flags under concurrent load
Shard: flags-memory
Idle process memory (RSS) 43.86 MiB 43.85 MiB <0.1% less memory Not flagged
Evaluate 1,000 flags under concurrent load
Shard: flags-memory
Observed peak process memory (RSS) 44.19 MiB 44.21 MiB <0.1% more memory Not flagged
Evaluate 1,000 flags under concurrent load
Shard: flags-memory
Process memory after settling (RSS) 44.19 MiB 44.21 MiB <0.1% more memory Not flagged
Evaluate 1,000 flags under concurrent load
Shard: flags-memory
Lifetime peak process memory (RSS) 44 MiB 44.12 MiB 0.3% more memory Not flagged
Refresh 1,000 flag definitions
Shard: flags-memory
Idle process memory (RSS) 43.01 MiB 43.02 MiB <0.1% more memory Not flagged
Refresh 1,000 flag definitions
Shard: flags-memory
Observed peak process memory (RSS) 43.02 MiB 43.03 MiB <0.1% more memory Not flagged
Refresh 1,000 flag definitions
Shard: flags-memory
Process memory after settling (RSS) 43.02 MiB 43.03 MiB <0.1% more memory Not flagged
Refresh 1,000 flag definitions
Shard: flags-memory
Lifetime peak process memory (RSS) 42.85 MiB 42.89 MiB 0.1% more memory Not flagged
Evaluate one boolean flag
Shard: flags-single
Allocated memory per operation 15.74 KiB/op 15.74 KiB/op No observed change Not flagged
Evaluate one boolean flag
Shard: flags-single
Allocations per operation 162.77 count/op 162.77 count/op No observed change Not flagged
Evaluate one boolean flag
Shard: flags-single
Time per operation 8.24 µs/op 8.21 µs/op 0.4% less time Not flagged
Evaluate one boolean flag
Shard: flags-single
Process CPU time per operation 8.24 µs/op 8.21 µs/op 0.3% less time Not flagged
Evaluate a cohort-filtered flag
Shard: flags-single
Allocated memory per operation 24.6 KiB/op 24.6 KiB/op No observed change Not flagged
Evaluate a cohort-filtered flag
Shard: flags-single
Allocations per operation 239.28 count/op 239.28 count/op No observed change Not flagged
Evaluate a cohort-filtered flag
Shard: flags-single
Time per operation 11.5 µs/op 11.41 µs/op 0.8% less time Not flagged
Evaluate a cohort-filtered flag
Shard: flags-single
Process CPU time per operation 11.5 µs/op 11.41 µs/op 0.8% less time Not flagged
Evaluate a flag (early match)
Shard: flags-single
Allocated memory per operation 39.81 KiB/op 39.81 KiB/op No observed change Not flagged
Evaluate a flag (early match)
Shard: flags-single
Allocations per operation 388.27 count/op 388.27 count/op No observed change Not flagged
Evaluate a flag (early match)
Shard: flags-single
Time per operation 16.4 µs/op 16.26 µs/op 0.9% less time Not flagged
Evaluate a flag (early match)
Shard: flags-single
Process CPU time per operation 16.4 µs/op 16.26 µs/op 0.9% less time Not flagged
Evaluate a flag (late match)
Shard: flags-single
Allocated memory per operation 60.59 KiB/op 60.59 KiB/op No observed change Not flagged
Evaluate a flag (late match)
Shard: flags-single
Allocations per operation 584.27 count/op 584.27 count/op No observed change Not flagged
Evaluate a flag (late match)
Shard: flags-single
Time per operation 23.5 µs/op 23.43 µs/op 0.3% less time Not flagged
Evaluate a flag (late match)
Shard: flags-single
Process CPU time per operation 23.5 µs/op 23.43 µs/op 0.3% less time Not flagged
Evaluate a flag (no match)
Shard: flags-single
Allocated memory per operation 59.22 KiB/op 59.22 KiB/op No observed change Not flagged
Evaluate a flag (no match)
Shard: flags-single
Allocations per operation 567.77 count/op 567.77 count/op No observed change Not flagged
Evaluate a flag (no match)
Shard: flags-single
Time per operation 22.26 µs/op 22.16 µs/op 0.4% less time Not flagged
Evaluate a flag (no match)
Shard: flags-single
Process CPU time per operation 22.26 µs/op 22.16 µs/op 0.4% less time Not flagged
Evaluate a property-filtered flag
Shard: flags-single
Allocated memory per operation 19.04 KiB/op 19.04 KiB/op No observed change Not flagged
Evaluate a property-filtered flag
Shard: flags-single
Allocations per operation 192.27 count/op 192.27 count/op No observed change Not flagged
Evaluate a property-filtered flag
Shard: flags-single
Time per operation 9.26 µs/op 9.31 µs/op 0.6% more time Not flagged
Evaluate a property-filtered flag
Shard: flags-single
Process CPU time per operation 9.26 µs/op 9.31 µs/op 0.5% more time Not flagged
Evaluate a regex-filtered flag
Shard: flags-single
Allocated memory per operation 20.66 KiB/op 20.66 KiB/op No observed change Not flagged
Evaluate a regex-filtered flag
Shard: flags-single
Allocations per operation 197.77 count/op 197.77 count/op No observed change Not flagged
Evaluate a regex-filtered flag
Shard: flags-single
Time per operation 9.8 µs/op 9.84 µs/op 0.4% more time Not flagged
Evaluate a regex-filtered flag
Shard: flags-single
Process CPU time per operation 9.8 µs/op 9.84 µs/op 0.4% more time Not flagged
Evaluate a rollout flag
Shard: flags-single
Allocated memory per operation 14.89 KiB/op 14.89 KiB/op No observed change Not flagged
Evaluate a rollout flag
Shard: flags-single
Allocations per operation 152.77 count/op 152.77 count/op No observed change Not flagged
Evaluate a rollout flag
Shard: flags-single
Time per operation 7.72 µs/op 7.74 µs/op 0.3% more time Not flagged
Evaluate a rollout flag
Shard: flags-single
Process CPU time per operation 7.72 µs/op 7.74 µs/op 0.3% more time Not flagged
Evaluate a multivariate flag
Shard: flags-single
Allocated memory per operation 15.9 KiB/op 15.9 KiB/op No observed change Not flagged
Evaluate a multivariate flag
Shard: flags-single
Allocations per operation 166.77 count/op 166.77 count/op No observed change Not flagged
Evaluate a multivariate flag
Shard: flags-single
Time per operation 8.56 µs/op 8.54 µs/op 0.3% less time Not flagged
Evaluate a multivariate flag
Shard: flags-single
Process CPU time per operation 8.56 µs/op 8.54 µs/op 0.3% less time Not flagged

Values are medians of paired measurements. Not flagged does not mean unchanged. Unavailable means a measurement could not be provided, not zero. Estimated allocation measurements retain their instrumentation uncertainty. RSS measures the whole process, not just SDK allocations; observed peaks are sampled checkpoints. Paired ranges in the diagnostics describe variability, not confidence intervals.

Each shard uses its own runner; compare Before PR against With PR within a row, not across shards.

Compared 44ac220377d12cf33da00abc5d3505ed86363b64 → e768ac0aa886cfdb41c3c3df44b92617224ffdfc.

Full report and diagnostics

102 metric rows; 102 rows shown in all results. Omitted or shortened data remain in comparison.json, including per-runner environments, resource limits, measurement methods, samples and checkpoints.

…ing tool spans

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant