Add custom headers from the env var LLM_EXTRA_HEADERS#476
Open
Luyansi3 wants to merge 1 commit into
Open
Conversation
Contributor
Greptile SummaryThis PR adds support for passing custom HTTP headers to every LiteLLM request via a new
Confidence Score: 3/5Not safe to merge as-is — supplying a syntactically valid but non-object JSON value (e.g. a number or boolean) triggers an unhandled exception at startup instead of the designed error panel. Both
Important Files Changed
Prompt To Fix All With AIFix the following 3 code review issues. Work through them one at a time, proposing concise fixes.
---
### Issue 1 of 3
strix/interface/main.py:92
The `TypeError` raised when `parsed` is not a dict (line 91) is not caught by `except (json.JSONDecodeError, ValueError)`, because `TypeError` is not a subclass of `ValueError`. When `LLM_EXTRA_HEADERS` is set to a valid JSON non-object (e.g. `42`, `true`, `"string"`), the `TypeError` escapes the try/except entirely, bypasses the nice error panel, and surfaces as an unhandled exception traceback instead of calling `sys.exit(1)` with a user-friendly message.
```suggestion
except (json.JSONDecodeError, TypeError, ValueError) as e:
```
### Issue 2 of 3
strix/interface/main.py:99-102
The example shell command is missing single quotes around the JSON value. Without them, the shell will interpret the spaces inside `{}` as argument separators, so copy-pasting this command will produce an incorrect assignment (the variable will only receive `{x-my-header:` and the rest will be treated as separate shell words).
```suggestion
error_text.append(
"export LLM_EXTRA_HEADERS='{\"x-my-header\": \"value\"}'\n",
style="dim white",
)
```
### Issue 3 of 3
strix/config/config.py:233-238
When `parsed` is valid JSON but not a `dict`, the `TypeError` is raised inside the `try` block but is not caught by `except json.JSONDecodeError`. It propagates to every call site (`warm_up_llm`, `check_duplicate`, `_summarize_messages`, `LLMConfig.__init__`) as an unhandled exception. Raising `ValueError` instead (consistent with how `json.JSONDecodeError` is already handled) makes the contract uniform and lets callers handle one exception type.
```suggestion
if isinstance(parsed, dict):
extra_headers = {str(k): str(v) for k, v in parsed.items() if v is not None}
else:
raise ValueError("LLM_EXTRA_HEADERS must be a JSON object, got a non-dict value")
except json.JSONDecodeError as e:
raise ValueError(f"Invalid LLM_EXTRA_HEADERS JSON: {e}") from e
```
Reviews (1): Last reviewed commit: "Add custom headers from the env var LLM_..." | Re-trigger Greptile |
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
| parsed = json.loads(raw_headers) | ||
| if not isinstance(parsed, dict): | ||
| raise TypeError("LLM_EXTRA_HEADERS must be a JSON object, got a non-dict value") | ||
| except (json.JSONDecodeError, ValueError) as e: |
Contributor
There was a problem hiding this comment.
TypeError raised when parsed is not a dict (line 91) is not caught by except (json.JSONDecodeError, ValueError), because TypeError is not a subclass of ValueError. When LLM_EXTRA_HEADERS is set to a valid JSON non-object (e.g. 42, true, "string"), the TypeError escapes the try/except entirely, bypasses the nice error panel, and surfaces as an unhandled exception traceback instead of calling sys.exit(1) with a user-friendly message.
Suggested change
| except (json.JSONDecodeError, ValueError) as e: | |
| except (json.JSONDecodeError, TypeError, ValueError) as e: |
Prompt To Fix With AI
This is a comment left during a code review.
Path: strix/interface/main.py
Line: 92
Comment:
The `TypeError` raised when `parsed` is not a dict (line 91) is not caught by `except (json.JSONDecodeError, ValueError)`, because `TypeError` is not a subclass of `ValueError`. When `LLM_EXTRA_HEADERS` is set to a valid JSON non-object (e.g. `42`, `true`, `"string"`), the `TypeError` escapes the try/except entirely, bypasses the nice error panel, and surfaces as an unhandled exception traceback instead of calling `sys.exit(1)` with a user-friendly message.
```suggestion
except (json.JSONDecodeError, TypeError, ValueError) as e:
```
How can I resolve this? If you propose a fix, please make it concise.
Comment on lines
+99
to
+102
| error_text.append( | ||
| 'export LLM_EXTRA_HEADERS={"x-my-header": "value"}\n', | ||
| style="dim white", | ||
| ) |
Contributor
There was a problem hiding this comment.
{} as argument separators, so copy-pasting this command will produce an incorrect assignment (the variable will only receive {x-my-header: and the rest will be treated as separate shell words).
Suggested change
| error_text.append( | |
| 'export LLM_EXTRA_HEADERS={"x-my-header": "value"}\n', | |
| style="dim white", | |
| ) | |
| error_text.append( | |
| "export LLM_EXTRA_HEADERS='{\"x-my-header\": \"value\"}'\n", | |
| style="dim white", | |
| ) |
Prompt To Fix With AI
This is a comment left during a code review.
Path: strix/interface/main.py
Line: 99-102
Comment:
The example shell command is missing single quotes around the JSON value. Without them, the shell will interpret the spaces inside `{}` as argument separators, so copy-pasting this command will produce an incorrect assignment (the variable will only receive `{x-my-header:` and the rest will be treated as separate shell words).
```suggestion
error_text.append(
"export LLM_EXTRA_HEADERS='{\"x-my-header\": \"value\"}'\n",
style="dim white",
)
```
How can I resolve this? If you propose a fix, please make it concise.
Comment on lines
+233
to
+238
| if isinstance(parsed, dict): | ||
| extra_headers = {str(k): str(v) for k, v in parsed.items() if v is not None} | ||
| else: | ||
| raise TypeError("LLM_EXTRA_HEADERS must be a JSON object") | ||
| except json.JSONDecodeError as e: | ||
| raise ValueError(f"Invalid LLM_EXTRA_HEADERS JSON: {e}") from e |
Contributor
There was a problem hiding this comment.
parsed is valid JSON but not a dict, the TypeError is raised inside the try block but is not caught by except json.JSONDecodeError. It propagates to every call site (warm_up_llm, check_duplicate, _summarize_messages, LLMConfig.__init__) as an unhandled exception. Raising ValueError instead (consistent with how json.JSONDecodeError is already handled) makes the contract uniform and lets callers handle one exception type.
Suggested change
| if isinstance(parsed, dict): | |
| extra_headers = {str(k): str(v) for k, v in parsed.items() if v is not None} | |
| else: | |
| raise TypeError("LLM_EXTRA_HEADERS must be a JSON object") | |
| except json.JSONDecodeError as e: | |
| raise ValueError(f"Invalid LLM_EXTRA_HEADERS JSON: {e}") from e | |
| if isinstance(parsed, dict): | |
| extra_headers = {str(k): str(v) for k, v in parsed.items() if v is not None} | |
| else: | |
| raise ValueError("LLM_EXTRA_HEADERS must be a JSON object, got a non-dict value") | |
| except json.JSONDecodeError as e: | |
| raise ValueError(f"Invalid LLM_EXTRA_HEADERS JSON: {e}") from e |
Prompt To Fix With AI
This is a comment left during a code review.
Path: strix/config/config.py
Line: 233-238
Comment:
When `parsed` is valid JSON but not a `dict`, the `TypeError` is raised inside the `try` block but is not caught by `except json.JSONDecodeError`. It propagates to every call site (`warm_up_llm`, `check_duplicate`, `_summarize_messages`, `LLMConfig.__init__`) as an unhandled exception. Raising `ValueError` instead (consistent with how `json.JSONDecodeError` is already handled) makes the contract uniform and lets callers handle one exception type.
```suggestion
if isinstance(parsed, dict):
extra_headers = {str(k): str(v) for k, v in parsed.items() if v is not None}
else:
raise ValueError("LLM_EXTRA_HEADERS must be a JSON object, got a non-dict value")
except json.JSONDecodeError as e:
raise ValueError(f"Invalid LLM_EXTRA_HEADERS JSON: {e}") from e
```
How can I resolve this? If you propose a fix, please make it concise.There was a problem hiding this comment.
Pull request overview
This PR adds LLM_EXTRA_HEADERS so Strix can attach user-supplied HTTP headers to LiteLLM requests, which is useful for proxy/gateway setups that require additional credentials or custom routing headers.
Changes:
- Added
LLM_EXTRA_HEADERSto config resolution andLLMConfig. - Forwarded extra headers to warm-up, dedupe, memory compression, and main LLM request paths.
- Added startup validation and documentation for the new environment variable.
Reviewed changes
Copilot reviewed 7 out of 7 changed files in this pull request and generated 5 comments.
Show a summary per file
| File | Description |
|---|---|
strix/llm/memory_compressor.py |
Passes resolved extra headers into memory summarization requests. |
strix/llm/llm.py |
Includes extra_headers in main completion request arguments. |
strix/llm/dedupe.py |
Forwards extra headers in duplicate-check LLM calls. |
strix/llm/config.py |
Stores resolved extra_headers on LLMConfig. |
strix/interface/main.py |
Validates LLM_EXTRA_HEADERS at startup and uses it during LLM warm-up. |
strix/config/config.py |
Adds the new config key and extends resolve_llm_config() to parse/return headers. |
docs/advanced/configuration.mdx |
Documents the new LLM_EXTRA_HEADERS setting. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| try: | ||
| parsed = json.loads(raw_headers) | ||
| if not isinstance(parsed, dict): | ||
| raise TypeError("LLM_EXTRA_HEADERS must be a JSON object, got a non-dict value") |
| error_text.append(f"Error: {e}\n", style="white") | ||
| error_text.append("\nExample:\n", style="white") | ||
| error_text.append( | ||
| 'export LLM_EXTRA_HEADERS={"x-my-header": "value"}\n', |
Comment on lines
212
to
+214
| model = Config.get("strix_llm") | ||
| if not model: | ||
| return None, None, None | ||
| return None, None, None, None |
| if isinstance(parsed, dict): | ||
| extra_headers = {str(k): str(v) for k, v in parsed.items() if v is not None} | ||
| else: | ||
| raise TypeError("LLM_EXTRA_HEADERS must be a JSON object") |
Comment on lines
+233
to
+239
| if isinstance(parsed, dict): | ||
| extra_headers = {str(k): str(v) for k, v in parsed.items() if v is not None} | ||
| else: | ||
| raise TypeError("LLM_EXTRA_HEADERS must be a JSON object") | ||
| except json.JSONDecodeError as e: | ||
| raise ValueError(f"Invalid LLM_EXTRA_HEADERS JSON: {e}") from e | ||
|
|
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
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.
Summary
Add support for passing custom HTTP headers to every LiteLLM request via a new
LLM_EXTRA_HEADERSenvironment variable (JSON object). This enables custom API base / gateway setups that require multiple credentials (e.g. gateway key + model key) or additional enterprise headers.Fixes #466
What changed
LLM_EXTRA_HEADERS(string) added to Strix config (Config.llm_extra_headers).strix/interface/main.pynow validatesLLM_EXTRA_HEADERSat startup and shows a clear error panel with an example when invalid.resolve_llm_config()now returns(model, api_key, api_base, extra_headers)and parsesLLM_EXTRA_HEADERSas JSON (must be an object).extra_headersis forwarded tolitellm.completion()in:warm_up_llm)check_duplicate)_summarize_messages)strix/llm/llm.py) viaLLMConfig.extra_headersdocs/advanced/configuration.mdxdocumentsLLM_EXTRA_HEADERSwith a JSON example.Usage
Set
LLM_EXTRA_HEADERSto a JSON object of headers:Notes
Testing