LCORE-433: Updated OpenAPI doc#384
Conversation
WalkthroughUpdates OpenAPI documentation for POST /v1/feedback: revises response codes/descriptions, replaces 400 with 401, updates 403 text, adds 500 referencing a new ErrorResponse model, and adjusts FeedbackRequest.categories type. Mirrors these changes in generated output docs. Changes
Sequence Diagram(s)sequenceDiagram
participant Client
participant API as /v1/feedback
participant Auth
participant Store as Storage
Client->>API: POST feedback (FeedbackRequest)
API->>Auth: Validate credentials
alt Missing/invalid creds
Auth-->>API: Unauthorized
API-->>Client: 401 Unauthorized
else Authorized
API->>Store: Persist feedback
alt Success
Store-->>API: OK
API-->>Client: 200 Feedback received and stored
else Permission issue
Store-->>API: Forbidden
API-->>Client: 403 Forbidden
else Internal error
Store-->>API: Error
API-->>Client: 500 ErrorResponse{detail}
end
end
Estimated code review effort🎯 2 (Simple) | ⏱️ ~8 minutes Possibly related PRs
Poem
✨ Finishing Touches🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
CodeRabbit Configuration File (
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 5
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
docs/openapi.md(3 hunks)docs/output.md(2 hunks)
🧰 Additional context used
🪛 LanguageTool
docs/output.md
[style] ~174-~174: Unless you want to emphasize “not”, use “cannot” which is more common.
Context: ...iddenresponse) | | 500 | User feedback can not be stored | [ErrorResponse](#errorrespo...
(CAN_NOT_PREMIUM)
docs/openapi.md
[style] ~174-~174: Unless you want to emphasize “not”, use “cannot” which is more common.
Context: ...iddenresponse) | | 500 | User feedback can not be stored | [ErrorResponse](#errorrespo...
(CAN_NOT_PREMIUM)
🪛 markdownlint-cli2 (0.17.2)
docs/output.md
168-168: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
169-169: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
170-170: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
171-171: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
172-172: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
173-173: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
174-174: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
docs/openapi.md
168-168: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
169-169: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
170-170: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
171-171: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
172-172: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
173-173: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
174-174: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: e2e_tests
| | 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse) | ||
| | | ||
| | 400 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) | ||
| | 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) | ||
| | | ||
| | 403 | User is not authorized | [ForbiddenResponse](#forbiddenresponse) | ||
| | 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse) | ||
| | | ||
| | 500 | User feedback can not be stored | [ErrorResponse](#errorresponse) | ||
| | |
There was a problem hiding this comment.
💡 Verification agent
🧩 Analysis chain
Fix response table formatting and "cannot" wording; verify 401 standardization.
- Rows need trailing pipes; remove stray “ |” lines (markdownlint MD055/MD056).
- Prefer “cannot” over “can not” (LanguageTool).
- If moving to 401 for credential issues is intentional across the API, consider following up on other endpoints in a separate PR.
-| 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse)
- |
-| 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse)
- |
-| 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse)
- |
-| 500 | User feedback can not be stored | [ErrorResponse](#errorresponse)
- |
+| 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse) |
+| 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) |
+| 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse) |
+| 500 | User feedback cannot be stored | [ErrorResponse](#errorresponse) |Optional verification:
🏁 Script executed:
#!/bin/bash
# Style/wording
rg -n "can not be stored|can not" docs
# Response code consistency for auth errors across docs
rg -n $'\\| 400 \\| Missing or invalid credentials provided by client' docs
rg -n $'\\| 401 \\| Missing or invalid credentials provided by client' docsLength of output: 1657
Update table formatting, wording, and verify 401 usage
- In docs/openapi.md (lines 168–175), add trailing
|on each row and remove stray blank-pipe lines. - Change “can not” → “cannot” for consistency (also appears in docs/output.md:174, docs/openapi.json:335, docs/deployment_guide.md:1125).
- This endpoint now returns 401 for auth errors, but other endpoints still list 400 (docs/openapi.md:89 & 329; docs/output.md:89 & 329). If you intend to standardize on 401, update those in this PR or file a follow-up.
-| 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse)
-|
-| 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse)
-|
-| 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse)
-|
-| 500 | User feedback can not be stored | [ErrorResponse](#errorresponse)
-|
+| 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse) |
+| 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) |
+| 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse) |
+| 500 | User feedback cannot be stored | [ErrorResponse](#errorresponse) |📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| | 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse) | |
| | | |
| | 400 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) | |
| | 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) | |
| | | |
| | 403 | User is not authorized | [ForbiddenResponse](#forbiddenresponse) | |
| | 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse) | |
| | | |
| | 500 | User feedback can not be stored | [ErrorResponse](#errorresponse) | |
| | | |
| | 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse) | | |
| | 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) | | |
| | 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse) | | |
| | 500 | User feedback cannot be stored | [ErrorResponse](#errorresponse) | |
🧰 Tools
🪛 LanguageTool
[style] ~174-~174: Unless you want to emphasize “not”, use “cannot” which is more common.
Context: ...iddenresponse) | | 500 | User feedback can not be stored | [ErrorResponse](#errorrespo...
(CAN_NOT_PREMIUM)
🪛 markdownlint-cli2 (0.17.2)
168-168: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
169-169: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
170-170: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
171-171: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
172-172: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
173-173: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
174-174: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
175-175: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
🤖 Prompt for AI Agents
In docs/openapi.md around lines 168–175, the response table rows are missing
trailing pipe characters and contain stray blank-pipe lines; add a trailing '|'
to each row and remove any empty pipe-only lines, change "can not" to "cannot"
in this file (and also update docs/output.md:174, docs/openapi.json:335,
docs/deployment_guide.md:1125 to maintain consistency), and reconcile the
status-code usage by either updating other endpoints that list 400 to 401 in
this PR for consistency or file a follow-up task documenting that those
occurrences need to be changed.
| ## ErrorResponse | ||
|
|
||
|
|
||
| Model representing error response for query endpoint. | ||
|
|
||
|
|
There was a problem hiding this comment.
🛠️ Refactor suggestion
Make ErrorResponse description endpoint-agnostic.
It’s used by POST /v1/feedback (500). Current wording suggests it’s query-specific.
-Model representing error response for query endpoint.
+Model representing a generic error response payload used by multiple endpoints.📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| ## ErrorResponse | |
| Model representing error response for query endpoint. | |
| ## ErrorResponse | |
| Model representing a generic error response payload used by multiple endpoints. |
🤖 Prompt for AI Agents
In docs/openapi.md around lines 611 to 616, the ErrorResponse description is
currently phrased as "Model representing error response for query endpoint"
which implies it is query-specific; update the text to be endpoint-agnostic
(e.g., "Model representing a generic error response returned by API endpoints")
and ensure any parameter/examples or following sentences refer to generic errors
rather than a specific query endpoint so the same model clearly applies to POST
/v1/feedback (500) and other endpoints.
| | sentiment | | User sentiment, if provided must be -1 or 1 | | ||
| | user_feedback | | Feedback on the LLM response. | | ||
| | categories | array[FeedbackCategory] | List of feedback categories that describe issues with the LLM response (for negative feedback). | | ||
| | categories | | List of feedback categories that describe issues with the LLM response (for negative feedback). | |
There was a problem hiding this comment.
💡 Verification agent
🧩 Analysis chain
Re-type FeedbackRequest.categories or document allowed values.
The type is now unspecified while examples use an array of strings. If the enum was intentionally dropped:
- Set type to array and clarify allowed string values in description; or
- Keep enum by referencing FeedbackCategory.
-| categories | | List of feedback categories that describe issues with the LLM response (for negative feedback). |
+| categories | array | List of feedback categories (strings) describing issues with the LLM response (for negative feedback). Allowed values: "incomplete", "not_relevant", ... |Follow-up: If you intend to retain FeedbackCategory as a component, consider restoring:
- categories: array[FeedbackCategory], and
- populate FeedbackCategory with the allowed enum values.
🏁 Script executed:
#!/bin/bash
# Locate FeedbackRequest definition in docs/openapi.md
rg -C3 "FeedbackRequest" docs/openapi.md || echo "FeedbackRequest not found"
# Find occurrences of FeedbackCategory
rg -C3 "FeedbackCategory" docs/openapi.md || echo "FeedbackCategory not found"Length of output: 855
🏁 Script executed:
#!/bin/bash
# Display the enum values defined in FeedbackCategory
rg -A10 "## FeedbackCategory" docs/openapi.mdLength of output: 362
Document FeedbackRequest.categories type and allowed values
The categories field currently lacks a type and doesn’t reference the FeedbackCategory enum or list allowed strings. Please update the schema so clients know what to send. For example:
• In docs/openapi.md under FeedbackRequest:
-| categories | | List of feedback categories that describe issues with the LLM response (for negative feedback). |
+| categories | array[string] | List of feedback categories describing issues with the LLM response (for negative feedback). Refer to `FeedbackCategory` for allowed values. |• Or inline the enum values if you’ve dropped the component, e.g.:
-| categories | | … |
+| categories | array[string] | … Allowed values: `"incomplete"`, `"not_relevant"`, `"incorrect"`, … |If you intend to keep the FeedbackCategory component, restore the reference:
categories:
type: array
items:
$ref: '#/components/schemas/FeedbackCategory'🤖 Prompt for AI Agents
In docs/openapi.md around line 677, the FeedbackRequest.categories entry is
missing a type and allowed values; update the schema to declare categories as an
array whose items reference the existing FeedbackCategory component or inline
the allowed enum strings. Specifically, replace the current empty cell with a
YAML schema block that sets type: array and items: $ref:
'#/components/schemas/FeedbackCategory' (or items: { type: string, enum: [list,
of, allowed, values] } if you removed the component), so clients know the
expected type and permitted values.
| | 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse) | ||
| | | ||
| | 400 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) | ||
| | 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) | ||
| | | ||
| | 403 | User is not authorized | [ForbiddenResponse](#forbiddenresponse) | ||
| | 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse) | ||
| | | ||
| | 500 | User feedback can not be stored | [ErrorResponse](#errorresponse) | ||
| | |
There was a problem hiding this comment.
💡 Verification agent
🧩 Analysis chain
Fix response table formatting and "cannot" wording; consider 401 consistency.
- The table rows are missing trailing pipes and have an extra standalone " |" line, triggering markdownlint MD055/MD056.
- Prefer “cannot” over “can not” (LanguageTool hint).
- Optional: You’ve standardized this endpoint to use 401; if that’s a global policy, consider aligning other endpoints later.
Apply:
-| 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse)
- |
-| 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse)
- |
-| 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse)
- |
-| 500 | User feedback can not be stored | [ErrorResponse](#errorresponse)
- |
+| 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse) |
+| 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) |
+| 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse) |
+| 500 | User feedback cannot be stored | [ErrorResponse](#errorresponse) |If you want to check for other occurrences:
🏁 Script executed:
#!/bin/bash
# Style/wording
rg -n "can not be stored|can not" docs
# Response code consistency for auth errors
rg -n $'\\| 400 \\| Missing or invalid credentials provided by client' docs
rg -n $'\\| 401 \\| Missing or invalid credentials provided by client' docsLength of output: 1657
Fix table formatting and wording in feedback endpoint docs
- Add missing trailing pipes to each row in docs/output.md (lines 168–175) and remove the extra standalone
|lines to satisfy MD055/MD056. - Replace “can not” with “cannot” in docs/output.md (line 174). This wording also appears in:
- docs/openapi.md (line 174)
- docs/openapi.json (description at line 335)
- Optional: “Missing or invalid credentials provided by client” is still documented under both 400 and 401 across the docs (output.md: 89, 170, 329; openapi.md: 89, 170, 329). If you’re standardizing on 401 globally, consider updating those 400 entries.
Apply to docs/output.md:
-| 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse)
- |
-| 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse)
- |
-| 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse)
- |
-| 500 | User feedback can not be stored | [ErrorResponse](#errorresponse)
- |
+| 200 | Feedback received and stored | [FeedbackResponse](#feedbackresponse) |
+| 401 | Missing or invalid credentials provided by client | [UnauthorizedResponse](#unauthorizedresponse) |
+| 403 | Client does not have permission to access resource | [ForbiddenResponse](#forbiddenresponse) |
+| 500 | User feedback cannot be stored | [ErrorResponse](#errorresponse) |And update the “can not” occurrences elsewhere:
rg -n "can not be stored" docs/openapi.md docs/openapi.json🧰 Tools
🪛 LanguageTool
[style] ~174-~174: Unless you want to emphasize “not”, use “cannot” which is more common.
Context: ...iddenresponse) | | 500 | User feedback can not be stored | [ErrorResponse](#errorrespo...
(CAN_NOT_PREMIUM)
🪛 markdownlint-cli2 (0.17.2)
168-168: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
169-169: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
170-170: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
171-171: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
172-172: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
173-173: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
174-174: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
175-175: Table column count
Expected: 3; Actual: 1; Too few cells, row will be missing data
(MD056, table-column-count)
🤖 Prompt for AI Agents
In docs/output.md around lines 168–175, fix the markdown table by adding the
missing trailing pipe character to each row and remove any stray standalone "|"
lines so every row is properly pipe-delimited; on the same lines replace the
phrase "can not" with "cannot". Also update the identical "can not be stored"
occurrences in docs/openapi.md (line ~174) and docs/openapi.json (description
around line ~335) to "cannot be stored". Optionally review and standardize the
status-message duplication where "Missing or invalid credentials provided by
client" appears under both 400 and 401 (files/lines noted in the comment) and,
if standardizing to 401, change those 400 entries accordingly.
| ## ErrorResponse | ||
|
|
||
|
|
||
| Model representing error response for query endpoint. | ||
|
|
||
|
|
There was a problem hiding this comment.
🛠️ Refactor suggestion
Broaden ErrorResponse description beyond “query endpoint.”
ErrorResponse is referenced by the feedback endpoint (500). The current description reads as query-specific and may confuse readers.
-Model representing error response for query endpoint.
+Model representing a generic error response payload used by multiple endpoints.📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| ## ErrorResponse | |
| Model representing error response for query endpoint. | |
| ## ErrorResponse | |
| Model representing a generic error response payload used by multiple endpoints. |
🤖 Prompt for AI Agents
In docs/output.md around lines 611 to 616, the ErrorResponse description is
narrowly labeled as "Model representing error response for query endpoint."
Update the text to a broader, endpoint-agnostic description (e.g., "Model
representing a generic error response returned by API endpoints") so it reflects
usage by other endpoints like feedback (500); ensure the wording indicates it is
used for server/error responses across endpoints rather than only the query
endpoint.
1 participant
Description
Updated OpenAPI doc
Type of change
Summary by CodeRabbit