core.response_utils

Response cleaning and JSON extraction utilities.

Uses json_repair for fault-tolerant parsing of LLM output that may contain trailing commas, unquoted keys, or other common JSON mistakes.

Response cleaning and JSON extraction utilities.

Uses json_repair for fault-tolerant parsing of LLM output that may contain trailing commas, unquoted keys, or other common JSON mistakes.

extract_json(text)

Extract JSON from text using fault-tolerant json_repair.

Handles markdown code blocks, trailing commas, unquoted keys, and other common LLM JSON mistakes.

Parameters:

text (str) – Response text that may contain JSON.

Returns:

Parsed JSON object or None if no valid JSON found.

Return type:

Any | None

clean_response(response)

Process and validate a response, removing think tags and extracting JSON.

Parameters:

response (Any) – The raw response from the AI client.

Returns:

Cleaned response. If JSON is detected, returns the parsed object. Otherwise returns the cleaned string with think tags removed.

Return type:

Any

Functions

clean_response(response)

Process and validate a response, removing think tags and extracting JSON.

Parameters:

response (Any) – The raw response from the AI client.

Returns:

Cleaned response. If JSON is detected, returns the parsed object. Otherwise returns the cleaned string with think tags removed.

Return type:

Any

extract_json(text)

Extract JSON from text using fault-tolerant json_repair.

Handles markdown code blocks, trailing commas, unquoted keys, and other common LLM JSON mistakes.

Parameters:

text (str) – Response text that may contain JSON.

Returns:

Parsed JSON object or None if no valid JSON found.

Return type:

Any | None