Implement Query Analyzer for NestBot#4303
Open
rudransh-shrivastava wants to merge 9 commits intoOWASP:feature/nestbot-ai-assistantfrom
Open
Implement Query Analyzer for NestBot#4303rudransh-shrivastava wants to merge 9 commits intoOWASP:feature/nestbot-ai-assistantfrom
rudransh-shrivastava wants to merge 9 commits intoOWASP:feature/nestbot-ai-assistantfrom
Conversation
2 tasks
Contributor
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
WalkthroughAdds a new AI-driven Query Analyzer module plus Jinja templates and integrates a single analysis call into the assistant flow; the assistant now logs the analysis (is_simple and sub_queries) but routing/control flow remain unchanged in this diff. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~22 minutes 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
📝 Coding Plan
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. Comment |
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
3 issues found across 5 files
Confidence score: 3/5
- There is concrete medium risk here:
backend/apps/ai/flows/assistant.pynow makes an unconditional LLM analysis call for every query, which is likely to increase latency and cost without current functional upside. - The most severe correctness concern is in
backend/apps/ai/templates/query_analyzer/tasks/analyze.jinja, where rawquerytext is not clearly delimited as data, so prompt-injection content can steer analyzer output. backend/apps/ai/query_analyzer.pyappears fragile when parsing LLM output and can silently default tois_simple=True, which may mask format drift and cause misclassification once this output is relied on.- Pay close attention to
backend/apps/ai/templates/query_analyzer/tasks/analyze.jinja,backend/apps/ai/flows/assistant.py, andbackend/apps/ai/query_analyzer.py- prompt safety, unconditional runtime cost, and brittle parsing/default behavior are the main merge risks.
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="backend/apps/ai/templates/query_analyzer/tasks/analyze.jinja">
<violation number="1" location="backend/apps/ai/templates/query_analyzer/tasks/analyze.jinja:1">
P1: User input is injected into the prompt without delimiting it as data, allowing prompt-injection instructions inside `query` to alter analyzer output.</violation>
</file>
<file name="backend/apps/ai/flows/assistant.py">
<violation number="1" location="backend/apps/ai/flows/assistant.py:109">
P1: This adds an unconditional LLM-based query analysis call, but its output is not used in control flow yet. It increases request latency/cost for every query with no functional benefit.</violation>
</file>
<file name="backend/apps/ai/query_analyzer.py">
<violation number="1" location="backend/apps/ai/query_analyzer.py:72">
P1: The analyzer output parsing is brittle and silently falls back to `is_simple=True` when the LLM format changes. Parse structured output (JSON) with validation before using defaults.</violation>
</file>
Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (3)
backend/apps/ai/flows/assistant.py (1)
110-110: Uselogger.infoinstead oflogger.warningfor successful analysis.
logger.warningis typically reserved for unexpected conditions or potential problems. A successful query analysis is a normal operational event and should useinfoordebuglevel.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/apps/ai/flows/assistant.py` at line 110, Change the log level from warning to info for successful query analysis by replacing the logger.warning call with logger.info where the successful analysis is logged (the logger.warning(...) invocation in assistant.py), so normal operational events use info instead of warning; keep the same message text and context so only the severity is changed.backend/apps/ai/query_analyzer.py (2)
86-89: Consider validatingrequired_agentsagainst known agent names.The parsed
required_agentsvalues are not validated againstAGENT_DESCRIPTIONS.keys(). If the LLM returns an invalid or misspelled agent name, downstream code may fail or behave unexpectedly when these results are used for orchestration.♻️ Proposed validation
elif line_lower.startswith("requiredagents:"): value = line.split(":", 1)[1].strip().lower() if value and value != "none": - required_agents = [a.strip() for a in value.split(",") if a.strip()] + parsed_agents = [a.strip() for a in value.split(",") if a.strip()] + valid_agents = set(AGENT_DESCRIPTIONS.keys()) + required_agents = [a for a in parsed_agents if a in valid_agents]🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/apps/ai/query_analyzer.py` around lines 86 - 89, Validate the parsed required_agents after the split by checking each normalized entry against the known agent names in AGENT_DESCRIPTIONS.keys(); replace the current assignment to required_agents with a filtered list that only includes valid agent names (normalizing case or mapping as needed), and log or warn about any invalid/misspelled entries (or raise a clear error) so downstream orchestration code uses only recognized agents; reference the required_agents variable and AGENT_DESCRIPTIONS.keys() in your change and perform the validation immediately after the current parsing block in query_analyzer.py.
41-95: Add logging for observability of the analysis workflow.The function performs an LLM call but doesn't log the raw result or parsing outcomes. This makes debugging difficult when the LLM returns unexpected output formats.
📝 Proposed logging additions
+import logging + +logger = logging.getLogger(__name__) + def analyze_query(query: str) -> dict: ... result = crew.kickoff() result_str = str(result) + logger.debug("Raw analysis result: %s", result_str) is_simple = True ... + logger.debug( + "Parsed analysis: is_simple=%s, sub_queries=%s, required_agents=%s", + is_simple, + sub_queries, + required_agents, + ) return {🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/apps/ai/query_analyzer.py` around lines 41 - 95, In analyze_query, add observability by logging the LLM/crew outputs and parsing outcomes: after result = crew.kickoff() log the raw result (result and/or result_str) for debugging, then after the parsing loop log the derived values is_simple, sub_queries, and required_agents and any parse anomalies (e.g., unexpected formats or missing fields); use the module logger (or existing logger instance) to emit at INFO/DEBUG for normal values and WARN/ERROR for parse failures so you can trace both the raw LLM response and the final parsed state.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@backend/apps/ai/flows/assistant.py`:
- Around line 108-118: The call to analyze_query(query) can raise and currently
sits unprotected, so wrap the analyze_query invocation and the subsequent
logger.warning usage in its own try/except block to prevent analysis failures
from aborting the main flow; on exception, catch and log the error (e.g., via
logger.exception or logger.warning with the exception) and set a safe default
for query_analysis (e.g., {"is_simple": False, "sub_queries": [],
"required_agents": []}) so the rest of assistant flow can continue using
query_analysis; update references to query_analysis in this scope (the
analyze_query call and the logger.warning block) accordingly.
In `@backend/apps/ai/query_analyzer.py`:
- Line 70: Wrap the call to crew.kickoff() in a try/except so failures (network,
rate limits, LLM timeouts, parse errors) don't propagate; on exception log the
error (including exception text) and return a safe default result (e.g., None,
empty list, or a Result object indicating failure) from the surrounding
function, and only re-raise if a caller needs to handle fatal cases; update the
line where result = crew.kickoff() to perform the guarded call and ensure
callers handle the chosen default.
---
Nitpick comments:
In `@backend/apps/ai/flows/assistant.py`:
- Line 110: Change the log level from warning to info for successful query
analysis by replacing the logger.warning call with logger.info where the
successful analysis is logged (the logger.warning(...) invocation in
assistant.py), so normal operational events use info instead of warning; keep
the same message text and context so only the severity is changed.
In `@backend/apps/ai/query_analyzer.py`:
- Around line 86-89: Validate the parsed required_agents after the split by
checking each normalized entry against the known agent names in
AGENT_DESCRIPTIONS.keys(); replace the current assignment to required_agents
with a filtered list that only includes valid agent names (normalizing case or
mapping as needed), and log or warn about any invalid/misspelled entries (or
raise a clear error) so downstream orchestration code uses only recognized
agents; reference the required_agents variable and AGENT_DESCRIPTIONS.keys() in
your change and perform the validation immediately after the current parsing
block in query_analyzer.py.
- Around line 41-95: In analyze_query, add observability by logging the LLM/crew
outputs and parsing outcomes: after result = crew.kickoff() log the raw result
(result and/or result_str) for debugging, then after the parsing loop log the
derived values is_simple, sub_queries, and required_agents and any parse
anomalies (e.g., unexpected formats or missing fields); use the module logger
(or existing logger instance) to emit at INFO/DEBUG for normal values and
WARN/ERROR for parse failures so you can trace both the raw LLM response and the
final parsed state.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 910919b4-cbe1-4069-8a49-454abd87cb94
📒 Files selected for processing (5)
backend/apps/ai/flows/assistant.pybackend/apps/ai/query_analyzer.pybackend/apps/ai/templates/query_analyzer/backstory.jinjabackend/apps/ai/templates/query_analyzer/tasks/analyze.jinjacspell/custom-dict.txt
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
♻️ Duplicate comments (1)
backend/apps/ai/query_analyzer.py (1)
79-80:⚠️ Potential issue | 🟠 MajorBroaden
crew.kickoff()failure handling beyond two exception types.At Line 79, catching only
RuntimeErrorandValueErroris narrow for LLM/network SDK failures; unhandled exceptions can still bubble and break request handling.🛡️ Proposed fix
- except (RuntimeError, ValueError): + except Exception: logger.exception("Query analysis failed for: %s", query) return { "is_simple": True, "sub_queries": [query], "required_agents": [], }For CrewAI (current stable), what exception classes can Crew.kickoff() raise during model/network/tool failures, and is catching only RuntimeError/ValueError considered sufficient?🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/apps/ai/query_analyzer.py` around lines 79 - 80, The except block for crew.kickoff() is too narrow—replace the current except (RuntimeError, ValueError): that wraps crew.kickoff() in query_analyzer.py with a broader except Exception: to catch all runtime/SDK/network failures (but avoid catching BaseException like KeyboardInterrupt/SystemExit); keep using logger.exception("Query analysis failed for: %s", query) so the full traceback and error details are logged, and ensure any necessary cleanup or controlled re-raise happens after the log if callers need to see the error.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@backend/apps/ai/query_analyzer.py`:
- Line 80: The analyzer is logging raw user queries (logger.exception("Query
analysis failed for: %s", query) and similar at lines 109-110); compute a
non-reversible fingerprint (e.g., SHA-256) of the query and the query length and
log those instead of the raw text. Update places that reference the query
variable in logging (e.g., logger.exception and any logger.error calls in this
module) to log something like "query_fingerprint=%s, query_length=%d" using the
hashed value and len(query); add an import for hashlib and a small helper (e.g.,
fingerprint_query(query)) in query_analyzer.py to produce the hex digest so you
don’t repeat hashing logic.
- Around line 98-102: The SubQueries parsing can yield an empty list for
delimiter-only or malformed values; inside the block that handles elif
line_lower.startswith("subqueries:") (where variables sub_queries and query are
used), after computing sub_queries = [q.strip() ...], add a fallback so that if
sub_queries is empty it is set to [query] to preserve the original single-query
execution plan instead of returning an empty list.
---
Duplicate comments:
In `@backend/apps/ai/query_analyzer.py`:
- Around line 79-80: The except block for crew.kickoff() is too narrow—replace
the current except (RuntimeError, ValueError): that wraps crew.kickoff() in
query_analyzer.py with a broader except Exception: to catch all
runtime/SDK/network failures (but avoid catching BaseException like
KeyboardInterrupt/SystemExit); keep using logger.exception("Query analysis
failed for: %s", query) so the full traceback and error details are logged, and
ensure any necessary cleanup or controlled re-raise happens after the log if
callers need to see the error.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 6bc725ff-79f4-4b30-9aec-119f72b1cd30
📒 Files selected for processing (1)
backend/apps/ai/query_analyzer.py
Contributor
There was a problem hiding this comment.
1 issue found across 1 file (changes from recent commits).
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="backend/apps/ai/query_analyzer.py">
<violation number="1" location="backend/apps/ai/query_analyzer.py:97">
P2: If the LLM returns malformed subquery output (e.g., `SubQueries: |` or `SubQueries: | |`), the list comprehension produces an empty list, overwriting the `[query]` default. This would result in an empty execution plan. Assign the parsed result to a temporary variable and only overwrite `sub_queries` if it's non-empty.</violation>
</file>
Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.
Contributor
There was a problem hiding this comment.
1 issue found across 1 file (changes from recent commits).
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="backend/apps/ai/query_analyzer.py">
<violation number="1" location="backend/apps/ai/query_analyzer.py:93">
P2: Add a test for the new fallback branch so `sub_queries` defaults to `[query]` when `SubQueries:` parses to an empty list.
(Based on your team's feedback about adding tests for new logic and bug fixes.) [FEEDBACK_USED]</violation>
</file>
Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## feature/nestbot-ai-assistant #4303 +/- ##
================================================================
- Coverage 99.38% 96.45% -2.94%
================================================================
Files 520 551 +31
Lines 16384 16987 +603
Branches 2238 2371 +133
================================================================
+ Hits 16284 16384 +100
- Misses 62 560 +498
- Partials 38 43 +5
Flags with carried forward coverage won't be shown. Click here to find out more.
... and 51 files with indirect coverage changes Continue to review full report in Codecov by Sentry.
🚀 New features to boost your workflow:
|
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (1)
backend/apps/ai/query_analyzer.py (1)
101-103:⚠️ Potential issue | 🟠 MajorAvoid logging raw query text (privacy leak).
Line 102 logs full user input. Please log a non-reversible fingerprint + length instead.
🔐 Proposed fix
+import hashlib import contextlib import logging @@ logger = logging.getLogger(__name__) + + +def _query_fingerprint(query: str) -> tuple[str, int]: + return hashlib.sha256(query.encode("utf-8")).hexdigest()[:12], len(query) @@ if not required_agents: - logger.warning("Query analysis returned no valid agents for: %s", query) + query_hash, query_length = _query_fingerprint(query) + logger.warning( + "Query analysis returned no valid agents (query_hash=%s, query_length=%d)", + query_hash, + query_length, + )🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/apps/ai/query_analyzer.py` around lines 101 - 103, The warning currently logs the raw user query when no agents are returned; replace that to avoid leaking sensitive input by computing a non-reversible fingerprint (e.g., SHA-256 hex digest of the query bytes) and using len(query) for length, then call logger.warning with a message that includes only the fingerprint and length (not the raw query). Update the code around the logger.warning in query_analyzer.py where required_agents is checked (the block referencing required_agents and logger.warning) to compute the hash/fingerprint and log fingerprint + length instead of the query string.
🧹 Nitpick comments (1)
backend/apps/ai/query_analyzer.py (1)
84-99: Harden parsing before this drives orchestration.Lines 84-99 depend on free-form text patterns. This is fragile for LLM format drift and can silently return defaults. Prefer strict structured output (e.g., JSON schema) and validate fields before returning.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/apps/ai/query_analyzer.py` around lines 84 - 99, The current free-form parsing loop in query_analyzer.py that extracts is_simple, sub_queries, and required_agents from result_str is fragile; update Analyze/parse logic to require and validate a strict structured output (e.g., JSON with keys "isSimple","subQueries","requiredAgents"), attempt to parse result_str as JSON first, validate types and values (booleans, list of strings), map/whitelist requiredAgents against AGENT_DESCRIPTIONS, and only fall back to the existing line-based parsing if JSON parse fails while logging a warning; ensure functions/variables referenced (is_simple, sub_queries, required_agents, AGENT_DESCRIPTIONS, and the loop that inspects result_str) perform explicit sanity checks and return clear errors or defaults instead of silently accepting malformed input.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Duplicate comments:
In `@backend/apps/ai/query_analyzer.py`:
- Around line 101-103: The warning currently logs the raw user query when no
agents are returned; replace that to avoid leaking sensitive input by computing
a non-reversible fingerprint (e.g., SHA-256 hex digest of the query bytes) and
using len(query) for length, then call logger.warning with a message that
includes only the fingerprint and length (not the raw query). Update the code
around the logger.warning in query_analyzer.py where required_agents is checked
(the block referencing required_agents and logger.warning) to compute the
hash/fingerprint and log fingerprint + length instead of the query string.
---
Nitpick comments:
In `@backend/apps/ai/query_analyzer.py`:
- Around line 84-99: The current free-form parsing loop in query_analyzer.py
that extracts is_simple, sub_queries, and required_agents from result_str is
fragile; update Analyze/parse logic to require and validate a strict structured
output (e.g., JSON with keys "isSimple","subQueries","requiredAgents"), attempt
to parse result_str as JSON first, validate types and values (booleans, list of
strings), map/whitelist requiredAgents against AGENT_DESCRIPTIONS, and only fall
back to the existing line-based parsing if JSON parse fails while logging a
warning; ensure functions/variables referenced (is_simple, sub_queries,
required_agents, AGENT_DESCRIPTIONS, and the loop that inspects result_str)
perform explicit sanity checks and return clear errors or defaults instead of
silently accepting malformed input.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: b86e6e46-e119-4516-bd41-3de88abe5210
📒 Files selected for processing (2)
backend/apps/ai/query_analyzer.pycspell/custom-dict.txt
✅ Files skipped from review due to trivial changes (1)
- cspell/custom-dict.txt
Collaborator
Author
|
Updating the query analyzer output to be cleaner. |
Contributor
There was a problem hiding this comment.
2 issues found across 3 files (changes from recent commits).
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="backend/apps/ai/templates/query_analyzer/tasks/analyze.jinja">
<violation number="1" location="backend/apps/ai/templates/query_analyzer/tasks/analyze.jinja:20">
P2: Clarify the `SubQuery` format for simple queries and exact agent names. As written, the model can omit the line for simple queries or return labels the parser rejects, which makes the analyzer fall back to `rag`.</violation>
</file>
<file name="backend/apps/ai/query_analyzer.py">
<violation number="1" location="backend/apps/ai/query_analyzer.py:99">
P2: Don't force simple queries into a fabricated `rag` sub-query when the analyzer returns no `SubQuery:` lines.</violation>
</file>
Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (1)
backend/apps/ai/query_analyzer.py (1)
98-100:⚠️ Potential issue | 🟠 MajorAvoid logging raw user query text in warning logs.
Line 99 logs raw
query, which is a privacy/compliance leak risk. Log a fingerprint + length instead.🔐 Proposed fix
+import hashlib @@ def analyze_query(query: str) -> dict: @@ + query_hash = hashlib.sha256(query.encode("utf-8")).hexdigest()[:12] @@ if not sub_queries: - logger.warning("Query analysis returned no valid sub-queries for: %s", query) + logger.warning( + "Query analysis returned no valid sub-queries (query_hash=%s, query_length=%d)", + query_hash, + len(query), + ) sub_queries = [{"query": query, "agent": "rag"}]🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/apps/ai/query_analyzer.py` around lines 98 - 100, The warning in query_analyzer.py currently logs the raw user `query` when `sub_queries` is empty; instead compute a safe fingerprint and length and log those; replace the logger.warning call that references `query` (the branch that sets `sub_queries = [{"query": query, "agent": "rag"}]`) with code that computes a deterministic hash (e.g., SHA-256 hex digest of query.encode('utf-8')) and the character count (len(query or "")) and log something like "Query analysis returned no valid sub-queries for fingerprint=%s length=%d" with the fingerprint and length, handling None/empty queries gracefully.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@backend/apps/ai/query_analyzer.py`:
- Around line 83-97: The current parser in query_analyzer.py (looping over
result_str, using line_lower.startswith and only accepting inline "| agent:") is
brittle and misses outputs like separate "SubQueries:" / "RequiredAgents:" lines
or variants of casing/spacing; update the parsing logic to accept both
"subquery(s):" and "requiredagent(s):" (case/whitespace-insensitive), handle
both inline forms ("| agent:") and multi-line forms (capture a subquery line
into a temp variable and attach the next agent line if it matches
AGENT_DESCRIPTIONS), and ensure existing variables (is_simple, sub_queries,
AGENT_DESCRIPTIONS, result_str) are used; keep backwards-compatible behavior so
inline parsing still works and ignore malformed entries with contextlib.suppress
as before.
---
Duplicate comments:
In `@backend/apps/ai/query_analyzer.py`:
- Around line 98-100: The warning in query_analyzer.py currently logs the raw
user `query` when `sub_queries` is empty; instead compute a safe fingerprint and
length and log those; replace the logger.warning call that references `query`
(the branch that sets `sub_queries = [{"query": query, "agent": "rag"}]`) with
code that computes a deterministic hash (e.g., SHA-256 hex digest of
query.encode('utf-8')) and the character count (len(query or "")) and log
something like "Query analysis returned no valid sub-queries for fingerprint=%s
length=%d" with the fingerprint and length, handling None/empty queries
gracefully.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 011f1569-5b99-4469-8192-e10e0e2e1086
📒 Files selected for processing (3)
backend/apps/ai/flows/assistant.pybackend/apps/ai/query_analyzer.pybackend/apps/ai/templates/query_analyzer/tasks/analyze.jinja
✅ Files skipped from review due to trivial changes (1)
- backend/apps/ai/templates/query_analyzer/tasks/analyze.jinja
🚧 Files skipped from review as they are similar to previous changes (1)
- backend/apps/ai/flows/assistant.py
rudransh-shrivastava
dismissed stale reviews from coderabbitai[bot] and cubic-dev-ai[bot]
via
634b81e
rudransh-shrivastava
force-pushed
the
feature/nestbot-ai-assistant-work
branch
from
2b4e069 to
634b81e
Compare
|
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
1 participant
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.
Proposed change
Resolves #4302
Note: semgrep scan fails due to outdated branch dependencies.
Implement query analyzer for nestbot.
Checklist
make check-testlocally: all warnings addressed, tests passed