Refactor: Rename 'kata' to 'entity' - #33
Merged
visahak merged 6 commits intoJan 26, 2026
Merged
Conversation
📝 WalkthroughWalkthroughRenames public backend types from "Kata" → "Entity", updates logger names and default DB identifiers, adjusts client/MCP references, and hardens phoenix_sync message/token extraction with defensive parsing, payload summaries, and wider token-usage fallbacks. Changes
Sequence Diagram(s)sequenceDiagram
actor Span
participant PhoenixSync
participant Parser as _parse_content
participant Normalizer
participant TokenUsage
Span->>PhoenixSync: deliver span (inputs, outputs, gen_ai/llm fields)
PhoenixSync->>Parser: parse string-encoded items
Parser-->>PhoenixSync: parsed dicts or failure
PhoenixSync->>Normalizer: normalize messages, skip non-dicts (log payload summary)
PhoenixSync->>TokenUsage: extract tokens (check gen_ai.usage.*, llm.usage.*, llm.token_count.*)
Normalizer-->>PhoenixSync: normalized messages
PhoenixSync-->>Span: return extracted messages + token usage (or logger.exception on error)
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 3 | ❌ 2❌ Failed checks (2 warnings)
✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing touches
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Fix all issues with AI agents
In `@kaizen/schema/conflict_resolution.py`:
- Line 17: The module docstring in kaizen/schema/conflict_resolution.py contains
a grammatical error: change the phrase "Produced by the LLM, to be processed by
a entity backend." to use the correct article, e.g. "Produced by the LLM, to be
processed by an entity backend." Update the module-level docstring accordingly.
In `@kaizen/sync/phoenix_sync.py`:
- Around line 163-179: The fallback for plain string output.value was
accidentally removed so when self._parse_content returns a string output_msgs
stays None; restore the fallback by setting output_msgs to a single-message list
(e.g., [{"role": "assistant", "content": parsed_output or output_val}]) inside
the else branch of the output handling block in phoenix_sync.py (where
output_msgs, attrs.get("output.value") and self._parse_content are used), so
plain text outputs populate assistant context; keep it inside the try/except and
maintain existing logging/exception behavior.
🧹 Nitpick comments (2)
kaizen/frontend/mcp/mcp_server.py (1)
78-80: Consider usingjson.dumps()instead ofstr()for non-string content.When
message["content"]is a list or dict, usingstr()produces Python repr format (e.g.,{'key': 'value'}) rather than valid JSON. This may cause parsing issues if the stored content needs to be deserialized later.♻️ Suggested fix
Entity( type="trajectory", content=message["content"] if isinstance(message["content"], str) - else str(message["content"]), + else json.dumps(message["content"]),kaizen/backend/base.py (1)
42-49: Missing@abstractmethoddecorator onsearch_entities.Unlike all other methods in this ABC,
search_entitieslacks the@abstractmethoddecorator. This means subclasses can silently inherit the base implementation (which returnsNone) rather than being forced to provide their own implementation.♻️ Suggested fix
+ `@abstractmethod` def search_entities( self, namespace_id: str, query: str | None = None, filters: dict | None = None, limit: int = 10 ) -> list[RecordedEntity]: pass
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Fix all issues with AI agents
In `@kaizen/sync/phoenix_sync.py`:
- Around line 365-368: The total_tokens lookup uses the wrong attribute key;
update the fallback chain to use attrs.get("gen_ai.usage.total_tokens") instead
of attrs.get("llm.usage.total_tokens") so it matches the pattern used for
prompt_tokens and completion_tokens (keep the existing fallback to
attrs.get("llm.token_count.total")). Locate the dict building where attrs is
read (the block creating "prompt_tokens", "completion_tokens", "total_tokens")
and replace the incorrect key to ensure consistency.
🧹 Nitpick comments (3)
kaizen/sync/phoenix_sync.py (3)
135-136: Bareexceptclauses silently swallow all exceptions.These catch-all blocks hide parsing failures and make debugging difficult. Consider catching
Exceptionexplicitly and logging at debug level to aid troubleshooting without cluttering logs.Proposed fix
- except: - pass + except Exception: + logger.debug(f"Failed to parse input.value: {input_val[:100] if input_val else ''}")Apply similar pattern at lines 151-152:
- except: - pass + except Exception: + logger.debug(f"Failed to parse input message: {msg}")Also applies to: 151-152
186-187: Same bareexceptissue in output parsing.Apply the same pattern here for consistency and debuggability.
Proposed fix
except: - pass + except Exception: + logger.debug(f"Failed to parse output.value: {output_val[:100] if output_val else ''}")And at lines 198-199:
- except: - pass + except Exception: + logger.debug(f"Failed to parse output message: {msg}")Also applies to: 198-199
520-523: Uselogger.exceptionto capture traceback.When logging inside an exception handler,
logger.exception()automatically includes the stack trace, which aids debugging.Proposed fix
except Exception as e: error_msg = f"Error processing span {span_id}: {e}" - logger.error(error_msg) + logger.exception(error_msg) errors.append(error_msg)
visahak
force-pushed
the
refactor/rename-kata-to-entity
branch
from
January 26, 2026 15:38
8f5fccc to
5d77adf
Compare
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Fix all issues with AI agents
In `@kaizen/sync/phoenix_sync.py`:
- Around line 148-152: Replace the bare except/pass around calls to
self._parse_content with an explicit Exception handler that logs the error at
debug level before continuing; specifically, in the blocks where msg is a str
and you call self._parse_content (the occurrences around the existing msg
parsing and the similar block later), change the except to "except Exception as
e" and call the module/class logger (or self.logger) to debug.log the exception
and the offending msg payload so malformed payloads are visible while preserving
behavior.
- Line 522: In the except Exception as e block where logger.error(error_msg) is
called (in phoenix_sync.py), replace the error-only call with
logger.exception(error_msg) so the exception stack trace is captured; locate the
except block around the logger.error call and change the logging invocation to
logger.exception(error_msg) (no need to pass e explicitly) to preserve traceback
for triage.
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Fix all issues with AI agents
In `@kaizen/sync/phoenix_sync.py`:
- Around line 135-137: The debug logs currently emit full raw payloads (e.g.,
logger.debug in the except block that logs input_val and similar calls at the
other locations), risking PII leakage; update these to avoid printing raw
content by logging only the payload type and length or a short redacted preview
(e.g., show first N chars + "...") and/or gate full payload logging behind a
secure flag; locate uses of logger.debug that reference input_val or input.value
(and the similar calls around lines 148-153, 186-187, 195-199) and replace them
with a call to a small helper (e.g., redact_preview(payload) or
format_payload_summary(payload)) that returns type/length or a redacted snippet
before logging.
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Fix all issues with AI agents
In `@kaizen/sync/phoenix_sync.py`:
- Around line 381-385: The current usage dict uses the `or` operator with
`attrs.get(...)` which treats valid 0 values as falsy; update the three lookups
for prompt_tokens, completion_tokens, and total_tokens to prefer the first
non-None value instead of truthy-bytes (e.g., implement a small helper like
`_first_non_none` and call it with attrs.get("gen_ai.usage.prompt_tokens"),
attrs.get("llm.token_count.prompt"), attrs.get("llm.usage.prompt_tokens") for
prompt_tokens, similarly for completion and total) so that explicit 0 is
preserved; ensure you replace the `or` chains in the usage construction with
calls to the helper or explicit None checks using `is not None`.
🧹 Nitpick comments (1)
kaizen/sync/phoenix_sync.py (1)
200-204: Remove dead code comment.The commented
#passon line 202 is a leftover from when the fallback was disabled. It serves no purpose now and should be removed.Proposed fix
else: # Fallback for simple string output output_msgs = [{"role": "assistant", "content": output_val}] - `#pass` except Exception as e:
vinodmut
approved these changes
Jan 26, 2026
Merged
This was referenced Feb 19, 2026
4 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
This PR addresses issue #27 by renaming all references of "kata" to "entity" throughout the codebase to better align with the domain model.
Changes
BaseKataBackendto BaseEntityBackend.MilvusKataBackendto MilvusEntityBackend.FilesystemKataBackendto FilesystemEntityBackend.entities.milvus.dbentities.sqlite.dbBreaking Changes
entities.milvus.db/entities.sqlite.dbor migrate data if they wish to preserve existing content.Related Issue
Fixes #27
Please merge after PR#32
Summary by CodeRabbit
Bug Fixes
Refactoring
Chores
Tests
✏️ Tip: You can customize this high-level summary in your review settings.