2.3

Adjusted Signal

2.3

Raw Pattern Score

100%

Time Factor

2026-10-28

Last Push

7.8K

Stars

Rust

Language

681.6K

Lines of Code

1.5K

Files

1.1K

Pattern Hits

2026-10-31

Scan Date

0.02

HC Hit Rate

Risk Review · Payload v1

Use each vector to guide manual verification of the underlying findings.

Payload schema: 1.0.0 · Engine: 2.0.0 · Scoring: heuristic-density-v1 · Scan date: 2026-10-31.

Cognitive Ownership Risk 7.2/100 · LOW

Can maintainers explain intent, assumptions, and trade-offs? 917 finding(s).

Hallucinated Dependency & Supply Chain Risk 0.1/100 · LOW

Are imports declared and package/API names verified? 16 finding(s).

Agent Over-Engineering Ratio 5.4/100 · LOW

Are wrappers, guards, and repeated structures necessary? 526 finding(s).

PR Review Velocity Impact 11.8/100 · LOW

Which changes need focused human review? 1129 finding(s).

What These Metrics Mean

Adjusted Signal: Legacy pattern score normalized per 1,000 scanned lines and multiplied by the repository-level time factor. It is not an estimate of AI authorship.
Raw Pattern Score: Sum of severity- and context-weighted pattern findings before applying the time factor.
Time Factor: A coarse multiplier based on repository last-push metadata. It does not identify when individual code was written.
Pattern Hits: Count of configured pattern findings. Review each location and surrounding code before acting.
HC Hit Rate: High- and critical-severity findings per scanned file. Severity is a configured detector weight, not proof of a vulnerability or authorship.
Lines / Files: Scan coverage denominators used by the legacy normalized score.

Pattern-Signal History

This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.

Severity Breakdown

Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.

CRITICAL 0HIGH 36MEDIUM 202LOW 891

Directory Score Breakdown

This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.

Pattern Findings

The scanner identified 1129 distinct pattern matches across 18 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.

Over-Commented Block372 hits · 360 pts

Severity	File	Line	Snippet	Context
LOW	mistralrs-macros/README.md	1	# mistralrs-macros	COMMENT
LOW	mistralrs-macros/src/lib.rs	1	//! Proc macros for ergonomic tool definition in mistral.rs	COMMENT
LOW	mistralrs-macros/src/lib.rs	21	//! #[description = "The city name"]	COMMENT
LOW	mistralrs-macros/src/lib.rs	81	}	COMMENT
LOW	mistralrs-macros/src/lib.rs	261	.collect();	COMMENT
LOW	mistralrs-macros/src/lib.rs	321		COMMENT
LOW	mistralrs-mcp/src/client.rs	161		COMMENT
LOW	mistralrs-mcp/src/client.rs	181	/// # Example	COMMENT
LOW	mistralrs-mcp/src/client.rs	241		COMMENT
LOW	mistralrs-mcp/src/client.rs	261	/// - A callback function that executes the tool with timeout and concurrency controls	COMMENT
LOW	mistralrs-mcp/src/client.rs	281	&self.tool_callbacks_with_tools	COMMENT
LOW	mistralrs-mcp/src/client.rs	301	/// # Example	COMMENT
LOW	mistralrs-mcp/src/client.rs	321		COMMENT
LOW	mistralrs-mcp/src/client.rs	661	);	COMMENT
LOW	mistralrs-mcp/src/client.rs	681	/// let mut client = McpClient::new(config);	COMMENT
LOW	mistralrs-mcp/src/client.rs	701	self.tool_callbacks_with_tools.clear();	COMMENT
LOW	mistralrs-mcp/src/client.rs	741	///	COMMENT
LOW	mistralrs-mcp/src/client.rs	781	}	COMMENT
LOW	mistralrs-mcp/src/client.rs	801	/// # }	COMMENT
LOW	mistralrs-mcp/src/client.rs	821	/// // Add a new server dynamically	COMMENT
LOW	mistralrs-mcp/src/client.rs	861	/// Closes the connection and removes the server from the configuration.	COMMENT
LOW	mistralrs-mcp/src/client.rs	901	/// ```rust,no_run	COMMENT
LOW	mistralrs-mcp/src/client.rs	921	}	COMMENT
LOW	mistralrs-mcp/src/client.rs	941	/// ```	COMMENT
LOW	mistralrs-mcp/src/client.rs	961	/// # }	COMMENT
LOW	mistralrs-mcp/src/client.rs	1021		COMMENT
LOW	mistralrs-mcp/src/client.rs	1041	// ==================== Status / Convenience Methods ====================	COMMENT
LOW	mistralrs-mcp/src/client.rs	1061		COMMENT
LOW	mistralrs-mcp/src/client.rs	1081	}	COMMENT
LOW	mistralrs-mcp/src/client.rs	1121	/// Returns a vector of (server_id, resource) tuples.	COMMENT
LOW	mistralrs-mcp/src/transport.rs	61		COMMENT
LOW	mistralrs-mcp/src/transport.rs	81	/// - Error Handling: Comprehensive JSON-RPC and HTTP error handling	COMMENT
LOW	mistralrs-mcp/src/transport.rs	101	/// headers.insert("X-API-Version".to_string(), "v1".to_string());	COMMENT
LOW	mistralrs-mcp/src/transport.rs	121	/// - Standard JSON responses	COMMENT
LOW	mistralrs-mcp/src/transport.rs	141	/// # Returns	COMMENT
LOW	mistralrs-mcp/src/transport.rs	161	/// None	COMMENT
LOW	mistralrs-mcp/src/transport.rs	201	/// * `sse_text` - Raw SSE response text from the server	COMMENT
LOW	mistralrs-mcp/src/transport.rs	261	/// # Arguments	COMMENT
LOW	mistralrs-mcp/src/transport.rs	281	/// use mistralrs_mcp::transport::{HttpTransport, McpTransport};	COMMENT
LOW	mistralrs-mcp/src/transport.rs	341	///	COMMENT
LOW	mistralrs-mcp/src/transport.rs	401	/// - Process Isolation: Each MCP server runs in its own process for security	COMMENT
LOW	mistralrs-mcp/src/transport.rs	421	///	COMMENT
LOW	mistralrs-mcp/src/transport.rs	441	/// Some(env) // Environment variables	COMMENT
LOW	mistralrs-mcp/src/transport.rs	461	///	COMMENT
LOW	mistralrs-mcp/src/transport.rs	481	/// # Arguments	COMMENT
LOW	mistralrs-mcp/src/transport.rs	501	///	COMMENT
LOW	mistralrs-mcp/src/transport.rs	581	/// a line-delimited JSON request to the process stdin and reads the	COMMENT
LOW	mistralrs-mcp/src/transport.rs	601	///	COMMENT
LOW	mistralrs-mcp/src/transport.rs	661	///	COMMENT
LOW	mistralrs-mcp/src/transport.rs	681	///	COMMENT
LOW	mistralrs-mcp/src/transport.rs	721	/// - Bearer Token Support: Authentication via Authorization header during handshake	COMMENT
LOW	mistralrs-mcp/src/transport.rs	741	/// let mut headers = HashMap::new();	COMMENT
LOW	mistralrs-mcp/src/transport.rs	781	/// # Returns	COMMENT
LOW	mistralrs-mcp/src/transport.rs	801	/// headers.insert("Authorization".to_string(), "Bearer token123".to_string());	COMMENT
LOW	mistralrs-mcp/src/transport.rs	861	/// - Unique request ID generation for response correlation	COMMENT
LOW	mistralrs-mcp/src/transport.rs	881	///	COMMENT
LOW	mistralrs-mcp/src/transport.rs	961	/// Sends a WebSocket ping frame to test connection health	COMMENT
LOW	mistralrs-mcp/src/transport.rs	981	Ok(())	COMMENT
LOW	mistralrs-mcp/src/lib.rs	1	//! Model Context Protocol (MCP) Client Implementation	COMMENT
LOW	mistralrs-mcp/src/lib.rs	21	//!	COMMENT
312 more matches not shown…

Decorative Section Separators100 hits · 300 pts

Severity	File	Line	Snippet	Context
MEDIUM	docs/src/content/docs/examples/python/qwen3.md	42	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/python/qwen3.md	44	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/python/qwen3.md	67	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/python/qwen3.md	69	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/python/qwen3.md	94	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/python/qwen3.md	96	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/smollm3.md	59	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/smollm3.md	61	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/smollm3.md	82	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/smollm3.md	84	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/smollm3.md	109	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/smollm3.md	111	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/qwen3.md	59	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/qwen3.md	61	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/qwen3.md	82	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/qwen3.md	84	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/qwen3.md	107	# ------------------------------------------------------------------	COMMENT
MEDIUM	docs/src/content/docs/examples/server/qwen3.md	109	# ------------------------------------------------------------------	COMMENT
MEDIUM	mistralrs-quant/src/utils/ops.rs	819	// ────────────────────────────── ArgSort / Sort ────────────────────────────────	COMMENT
MEDIUM	mistralrs-quant/src/utils/ops.rs	4398	// ─────────────────────────────── Sort / ArgSort ────────────────────────────────	COMMENT
MEDIUM	mistralrs-quant/src/utils/ops.rs	4438	// ─────────────────────────────── 2 048-element vector ────────────────────────────────	COMMENT
MEDIUM	examples/python/qwen3.py	28	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/python/qwen3.py	30	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/python/qwen3.py	53	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/python/qwen3.py	55	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/python/qwen3.py	80	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/python/qwen3.py	82	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/qwen3.py	45	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/qwen3.py	47	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/qwen3.py	68	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/qwen3.py	70	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/qwen3.py	93	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/qwen3.py	95	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/smollm3.py	45	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/smollm3.py	47	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/smollm3.py	68	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/smollm3.py	70	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/smollm3.py	95	# ------------------------------------------------------------------	COMMENT
MEDIUM	examples/server/smollm3.py	97	# ------------------------------------------------------------------	COMMENT
MEDIUM	scripts/upload_wheels.py	23	# ============================================================================	COMMENT
MEDIUM	scripts/upload_wheels.py	25	# ============================================================================	COMMENT
MEDIUM	scripts/upload_wheels.py	51	# ============================================================================	COMMENT
MEDIUM	scripts/upload_wheels.py	53	# ============================================================================	COMMENT
MEDIUM	scripts/upload_wheels.py	106	# ============================================================================	COMMENT
MEDIUM	scripts/upload_wheels.py	108	# ============================================================================	COMMENT
MEDIUM	scripts/upload_wheels.py	184	# ============================================================================	COMMENT
MEDIUM	scripts/upload_wheels.py	186	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	28	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	30	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	76	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	78	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	203	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	205	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	252	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	254	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	278	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	280	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	423	# ============================================================================	COMMENT
MEDIUM	scripts/build_wheels.py	425	# ============================================================================	COMMENT
MEDIUM	…tralrs-core/src/pipeline/loaders/multimodal_loaders.rs	8804	// ─── Voxtral ────────────────────────────────────────────────────────────────	COMMENT
40 more matches not shown…

Modern AI Meta-Vocabulary70 hits · 210 pts

Severity	File	Line	Snippet	Context
MEDIUM	mistralrs-macros/src/lib.rs	4	//! that can be used with the mistral.rs agentic loop.	COMMENT
MEDIUM	mistralrs-macros/src/lib.rs	86	/// used with the mistral.rs agentic loop. It generates:	COMMENT
MEDIUM	mistralrs-mcp/src/tools.rs	135	/// Context provided to tool callbacks by the agentic loop.	COMMENT
MEDIUM	mistralrs-cli/webui/src/lib/stores/chat.svelte.ts	532	// Fork the agentic session so the edited branch diverges cleanly from the prior turns.	COMMENT
MEDIUM	mistralrs-cli/webui/src/lib/stores/chat.svelte.ts	581	// Fork the agentic session so branches have independent state. Count complete turns BEFORE the	COMMENT
MEDIUM	mistralrs-cli/src/ui/types.rs	58	/// Server-side agentic session ID. Full session lives in `chat_<id>.session.json`.	COMMENT
MEDIUM	mistralrs-cli/src/ui/handlers/api.rs	751	/// Export the agentic session to a sidecar file beside the chat. Stamps the session_id into the chat JSON.	COMMENT
MEDIUM	mistralrs-cli/src/args/server.rs	30	/// Default maximum tool-call rounds for the agentic loop.	COMMENT
MEDIUM	mistralrs-cli/src/args/mod.rs	724	/// Build a local agent: enables web search, Python code execution, and shell execution, runs the agentic	COMMENT
MEDIUM	mistralrs-cli/src/args/mod.rs	727	#[arg(long, alias = "agentic")]	COMMENT
MEDIUM	mistralrs-pyo3/src/lib.rs	2168	/// Export an agentic session as a JSON string. `None` if missing.	COMMENT
MEDIUM	mistralrs-pyo3/src/lib.rs	2187	/// Import an agentic session from a JSON string. Replaces any existing session with the same ID.	COMMENT
MEDIUM	mistralrs-pyo3/src/lib.rs	2202	/// Delete an agentic session. Returns whether the session existed.	COMMENT
MEDIUM	mistralrs-pyo3/src/lib.rs	2210	/// List all stored agentic session IDs.	COMMENT
MEDIUM	mistralrs-pyo3/src/requests.rs	348	/// Session ID for persistent agentic state across requests.	COMMENT
MEDIUM	…rc/content/docs/guides/serve/openai-compatible-apis.md	89	## Tools, structured output, and agentic features	COMMENT
MEDIUM	docs/src/content/docs/examples/python/agentic-tools.md	118	# max_tool_rounds enables the agentic loop: model calls tools,	COMMENT
MEDIUM	docs/src/content/docs/examples/rust/advanced/agent.md	23	//! Example demonstrating the agentic loop with tool calling	COMMENT
MEDIUM	docs/src/content/docs/examples/rust/advanced/agent.md	28	//! 3. Run the agentic loop (non-streaming)	COMMENT
MEDIUM	…content/docs/examples/rust/advanced/agent-streaming.md	24	//! Example demonstrating the agentic loop with streaming output	COMMENT
MEDIUM	…content/docs/examples/rust/advanced/agent-streaming.md	29	//! 3. Run the agentic loop with streaming output	COMMENT
MEDIUM	…src/content/docs/examples/rust/advanced/file-inputs.md	22	//! the prompt and can be paginated with the built-in file tools when the agentic	COMMENT
MEDIUM	…lrs-server-core/src/mistralrs_server_router_builder.rs	57	/// Server-level defaults for agentic features.	COMMENT
MEDIUM	…lrs-server-core/src/mistralrs_server_router_builder.rs	116	/// Server-level agentic defaults	COMMENT
MEDIUM	…lrs-server-core/src/mistralrs_server_router_builder.rs	201	/// Sets the default maximum tool-call rounds for the agentic loop.	COMMENT
MEDIUM	mistralrs-server-core/src/approvals.rs	183	/// Decision payload for a pending agentic tool approval.	COMMENT
MEDIUM	mistralrs-server-core/src/chat_completion.rs	78	/// produced by the agentic loop are still reachable via `GET /v1/files/{id}` but	COMMENT
MEDIUM	mistralrs-server-core/src/responses.rs	586	/// Maximum number of agentic tool rounds (mistral.rs extension)	COMMENT
MEDIUM	mistralrs-server-core/src/files.rs	62	/// Which agentic tool produced the file, and when in the session.	COMMENT
MEDIUM	mistralrs-server-core/src/streaming.rs	80	// Only token steps carry decoded output; agentic control events and done markers share the channel	COMMENT
MEDIUM	…tralrs-server-core/src/mistralrs_for_server_builder.rs	1502	/// Creates a search embedding model configuration for agentic search reranking.	COMMENT
MEDIUM	mistralrs-server-core/src/openai.rs	1202	/// Permission policy for agentic tools.	COMMENT
MEDIUM	mistralrs-server-core/src/openai.rs	1215	/// Persistent agentic state. If `None`, a new session is created and the ID is returned in the response.	COMMENT
MEDIUM	examples/python/agentic_tools.py	96	# max_tool_rounds enables the agentic loop: model calls tools,	COMMENT
MEDIUM	mistralrs-core/src/response.rs	191	/// Ordered record of all tool calls made during the agentic loop.	COMMENT
MEDIUM	mistralrs-core/src/response.rs	197	/// Reuse in later requests to keep agentic state across messages.	COMMENT
MEDIUM	mistralrs-core/src/response.rs	296	/// Tool-specific structured progress data for agentic tool calls.	COMMENT
MEDIUM	mistralrs-core/src/response.rs	331	/// Phase of an agentic tool call.	COMMENT
MEDIUM	mistralrs-core/src/response.rs	388	/// Progress event emitted by the agentic loop during tool execution.	COMMENT
MEDIUM	mistralrs-core/src/request.rs	438	/// When true, registered code-execution tools are injected and the agentic loop runs.	COMMENT
MEDIUM	mistralrs-core/src/request.rs	441	/// When true, registered shell tools are injected and the agentic loop runs.	COMMENT
MEDIUM	mistralrs-core/src/request.rs	464	/// Persistent agentic state. If `None`, a new session is created and the ID is returned in the response.	COMMENT
MEDIUM	mistralrs-core/src/lib.rs	1957	/// Export an agentic session by ID. Bundles the session's files (full bodies). `None` if missing.	COMMENT
MEDIUM	mistralrs-core/src/lib.rs	2021	/// Delete an agentic session. Returns whether the session existed.	COMMENT
MEDIUM	mistralrs-core/src/files/store.rs	11	/// Per-entry TTL. Matches the agentic session default.	COMMENT
MEDIUM	mistralrs-core/src/files/mod.rs	1	//! Typed file outputs from agentic runs.	COMMENT
MEDIUM	mistralrs-core/src/files/mod.rs	87	/// A file produced by an agentic run.	COMMENT
MEDIUM	mistralrs-core/src/files/inject.rs	1	//! Helpers for surfacing `File`s to the model and adding the required-files contract to agentic tools.	COMMENT
MEDIUM	mistralrs-core/src/engine/agentic_loop.rs	1295	// Prevent the inner probe from re-entering the agentic loop or being rejected	COMMENT
MEDIUM	mistralrs-core/src/engine/agentic_loop.rs	1296	// by the files-without-agentic-surface guard in `add_request`.	COMMENT
MEDIUM	mistralrs-core/src/engine/agentic_session.rs	17	/// A stored agentic conversation, tool call/response messages included.	COMMENT
MEDIUM	mistralrs/examples/advanced/file_inputs/main.rs	4	//! the prompt and can be paginated with the built-in file tools when the agentic	COMMENT
MEDIUM	mistralrs/examples/advanced/agent/main.rs	1	//! Example demonstrating the agentic loop with tool calling	COMMENT
MEDIUM	mistralrs/examples/advanced/agent/main.rs	6	//! 3. Run the agentic loop (non-streaming)	COMMENT
MEDIUM	mistralrs/examples/advanced/agent_streaming/main.rs	1	//! Example demonstrating the agentic loop with streaming output	COMMENT
MEDIUM	mistralrs/examples/advanced/agent_streaming/main.rs	6	//! 3. Run the agentic loop with streaming output	COMMENT
MEDIUM	mistralrs/src/agent.rs	3	//! This module provides an `Agent` that runs an agentic loop with tool calling.	COMMENT
MEDIUM	mistralrs/src/agent.rs	78	/// Configuration for the agentic loop	COMMENT
MEDIUM	mistralrs/src/agent.rs	471	/// An agent that runs an agentic loop with tool calling	COMMENT
MEDIUM	mistralrs/src/agent.rs	528	/// Run the agentic loop with the given user message	COMMENT
10 more matches not shown…

Hyper-Verbose Identifiers329 hits · 210 pts

Severity	File	Line	Snippet	Context
LOW	mistralrs-code-exec/python/executor.py	134	def capture_matplotlib_figures():	CODE
LOW	…rc/content/docs/examples/python/multi-model-example.md	49	def example_runner_with_model_id():	CODE
LOW	…rc/content/docs/examples/python/multi-model-example.md	154	def example_streaming_with_models():	CODE
LOW	…rc/content/docs/examples/python/multi-model-example.md	183	def example_multi_model_setup():	CODE
LOW	…s/src/content/docs/examples/python/test-multi-model.md	40	def test_multi_model_operations():	CODE
LOW	…s/src/content/docs/examples/python/test-multi-model.md	118	def test_model_id_in_requests():	CODE
LOW	…s/src/content/docs/examples/server/multi-model-chat.md	157	def model_specific_tasks_example():	CODE
LOW	…s/src/content/docs/examples/server/multi-model-chat.md	245	def interactive_model_selection():	CODE
LOW	examples/python/test_multi_model.py	21	def test_multi_model_operations():	CODE
LOW	examples/python/test_multi_model.py	99	def test_model_id_in_requests():	CODE
LOW	examples/python/multi_model_example.py	26	def example_runner_with_model_id():	CODE
LOW	examples/python/multi_model_example.py	131	def example_streaming_with_models():	CODE
LOW	examples/python/multi_model_example.py	160	def example_multi_model_setup():	CODE
LOW	examples/server/multi_model_chat.py	135	def model_specific_tasks_example():	CODE
LOW	examples/server/multi_model_chat.py	223	def interactive_model_selection():	CODE
LOW⚡	scripts/test_production_soak.py	757	def test_observed_active_blocks_reduce_pressure_concurrency(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	766	def test_cached_prefixes_remain_reclaimable_for_eviction_pressure(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	776	def test_configured_limits_cap_capacity_concurrency(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	785	def test_rejects_capacity_that_cannot_fit_one_request(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3054	def test_ordering_and_cross_concurrency_do_not_exact_gate(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3061	def test_ungated_exact_diagnostics_still_require_complete_coverage(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3071	def test_invariance_requires_every_replay_ordering_and_cross_phase(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3398	def test_post_admission_cancellations_fit_sequence_capacity(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3407	def test_acceptance_adversarial_requires_server_pid(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3414	def test_acceptance_adversarial_requires_sustained_churn(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3645	def test_invalid_sparse_fallback_ratio_is_rejected(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3653	def test_invalid_sparse_accounting_coverage_is_rejected(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3661	def test_production_memory_gate_defaults_validate(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3741	def test_invalid_resource_utilization_is_rejected(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3747	def test_acceptance_grade_requires_mtp_and_both_graph_components(self) -> None:	CODE
LOW⚡	scripts/test_production_soak.py	3757	def test_non_acceptance_calibration_can_disable_mtp_and_dflash(self) -> None:	CODE
LOW	scripts/test_production_soak.py	118	def test_exact_text_corrects_large_round_trip_boundary_drift(self) -> None:	CODE
LOW	scripts/test_production_soak.py	132	def test_exact_text_uses_bounded_padding_for_token_count_gaps(self) -> None:	CODE
LOW	scripts/test_production_soak.py	149	def test_exact_text_reuses_encoded_context_source(self) -> None:	CODE
LOW	scripts/test_production_soak.py	159	def test_exact_text_grows_source_after_cross_boundary_contraction(self) -> None:	CODE
LOW	scripts/test_production_soak.py	171	def test_coherent_restatement_does_not_trigger_loop_gate(self) -> None:	CODE
LOW	scripts/test_production_soak.py	200	def test_severe_repeated_motif_is_rejected(self) -> None:	CODE
LOW	scripts/test_production_soak.py	220	def test_tail_loop_after_coherent_prose_is_rejected(self) -> None:	CODE
LOW	scripts/test_production_soak.py	245	def test_one_coherent_repetition_is_allowed(self) -> None:	CODE
LOW	scripts/test_production_soak.py	265	def test_short_output_has_stable_degeneration_evidence(self) -> None:	CODE
LOW	scripts/test_production_soak.py	275	def test_empty_stream_delta_does_not_invent_a_token(self) -> None:	CODE
LOW	scripts/test_production_soak.py	290	def test_coalesced_chunks_use_streamed_token_weight(self) -> None:	CODE
LOW	scripts/test_production_soak.py	308	def test_stop_excludes_one_unstreamed_terminal_token(self) -> None:	CODE
LOW	scripts/test_production_soak.py	326	def test_length_keeps_full_reported_token_denominator(self) -> None:	CODE
LOW	scripts/test_production_soak.py	343	def test_production_windows_use_streamable_denominator(self) -> None:	CODE
LOW	scripts/test_production_soak.py	362	def test_timeline_rates_weight_coalesced_chunks_by_tokens(self) -> None:	CODE
LOW	scripts/test_production_soak.py	389	def test_aggregate_coverage_cannot_hide_a_bad_request(self) -> None:	CODE
LOW	scripts/test_production_soak.py	416	def test_latency_windows_use_first_token_time(self) -> None:	CODE
LOW	scripts/test_production_soak.py	511	async def test_read_timeout_closes_streaming_response(self) -> None:	CODE
LOW	scripts/test_production_soak.py	558	def test_complete_window_includes_inflight_request_tokens(self) -> None:	CODE
LOW	scripts/test_production_soak.py	598	def test_baseline_interval_excludes_worker_ramp(self) -> None:	CODE
LOW	scripts/test_production_soak.py	629	def test_replay_eligibility_is_block_aligned(self) -> None:	CODE
LOW	scripts/test_production_soak.py	635	def test_prefix_gate_uses_only_replay_eligible_tokens_for_reuse(self) -> None:	CODE
LOW	scripts/test_production_soak.py	663	def test_prefix_gate_rejects_vacuous_reuse_contract(self) -> None:	CODE
LOW	scripts/test_production_soak.py	689	def test_prefix_gate_rejects_missing_reuse(self) -> None:	CODE
LOW	scripts/test_production_soak.py	742	def test_active_concurrency_fits_observed_kv_capacity_with_headroom(self) -> None:	CODE
LOW	scripts/test_production_soak.py	850	def test_source_cases_reconstruct_production_rng_identity(self) -> None:	CODE
LOW	scripts/test_production_soak.py	981	def test_replay_cohort_pads_one_full_batch_without_new_prompt_identities(	CODE
LOW	scripts/test_production_soak.py	1051	def test_cold_prefix_evidence_accounts_for_duplicate_prompt_identities(	CODE
LOW	scripts/test_production_soak.py	1131	def test_missing_hit_counter_remains_a_failed_hit_expectation(self) -> None:	CODE
269 more matches not shown…

Cross-File Repetition22 hits · 110 pts

Severity	File	Snippet	Context
HIGH	docs/src/content/docs/examples/python/lfm2.md	liquidai lfm2.5 moe text generation with the python sdk.	STRING
HIGH	examples/python/lfm2.py	liquidai lfm2.5 moe text generation with the python sdk.	STRING
HIGH	docs/src/content/docs/examples/python/lfm2-moe.md	liquidai lfm2.5 moe text generation with the python sdk.	STRING
HIGH	examples/python/lfm2_moe.py	liquidai lfm2.5 moe text generation with the python sdk.	STRING
HIGH	docs/src/content/docs/examples/python/shell-skills.md	--- name: invoice-auditor description: checks invoice line items and totals with a local python helper. --- # invoice au	STRING
HIGH	docs/src/content/docs/examples/server/skills.md	--- name: invoice-auditor description: checks invoice line items and totals with a local python helper. --- # invoice au	STRING
HIGH	examples/python/shell_skills.py	--- name: invoice-auditor description: checks invoice line items and totals with a local python helper. --- # invoice au	STRING
HIGH	examples/server/skills.py	--- name: invoice-auditor description: checks invoice line items and totals with a local python helper. --- # invoice au	STRING
HIGH	docs/src/content/docs/examples/python/shell-skills.md	item,amount hosting,25.00 storage,12.50 support,17.50 declared_total,55.00	STRING
HIGH	…s/src/content/docs/examples/server/anthropic-skills.md	item,amount hosting,25.00 storage,12.50 support,17.50 declared_total,55.00	STRING
HIGH	docs/src/content/docs/examples/server/skills.md	item,amount hosting,25.00 storage,12.50 support,17.50 declared_total,55.00	STRING
HIGH	examples/python/shell_skills.py	item,amount hosting,25.00 storage,12.50 support,17.50 declared_total,55.00	STRING
HIGH	examples/server/skills.py	item,amount hosting,25.00 storage,12.50 support,17.50 declared_total,55.00	STRING
HIGH	examples/server/anthropic_skills.py	item,amount hosting,25.00 storage,12.50 support,17.50 declared_total,55.00	STRING
HIGH	docs/src/content/docs/examples/python/shell-skills.md	import csv import sys with open(sys.argv[1], newline="") as handle: rows = list(csv.dictreader(handle)) declared = float	STRING
HIGH	docs/src/content/docs/examples/server/skills.md	import csv import sys with open(sys.argv[1], newline="") as handle: rows = list(csv.dictreader(handle)) declared = float	STRING
HIGH	examples/python/shell_skills.py	import csv import sys with open(sys.argv[1], newline="") as handle: rows = list(csv.dictreader(handle)) declared = float	STRING
HIGH	examples/server/skills.py	import csv import sys with open(sys.argv[1], newline="") as handle: rows = list(csv.dictreader(handle)) declared = float	STRING
HIGH	docs/src/content/docs/examples/server/anthropic-chat.md	streaming anthropic messages api example for the mistral.rs server. run the server: mistralrs serve -p 1234 -m qwen/qwen	STRING
HIGH	examples/server/anthropic_chat.py	streaming anthropic messages api example for the mistral.rs server. run the server: mistralrs serve -p 1234 -m qwen/qwen	STRING
HIGH	…rc/content/docs/examples/server/anthropic-streaming.md	streaming anthropic messages api example for the mistral.rs server. run the server: mistralrs serve -p 1234 -m qwen/qwen	STRING
HIGH	examples/server/anthropic_streaming.py	streaming anthropic messages api example for the mistral.rs server. run the server: mistralrs serve -p 1234 -m qwen/qwen	STRING

Structural Annotation Overuse42 hits · 77 pts

Severity	File	Line	Snippet	Context
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	458	// Step 1: load LSE_partial from gmem -> smem	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	503	// Step 2: Load O_partial from gmem -> smem for split = 0, 1, ..., kStages - 2.	COMMENT
LOW⚡	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	564	// Step 3: load and transpose LSE_partial from smem -> rmem	COMMENT
LOW⚡	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	574	// Step 4: compute the final LSE along the split dimension	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	623	// Step 5: store final LSE back to gmem	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	646	// Step 6: read O_partial from gmem -> smem -> rmem and accumulate the final O	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	689	// Step 7: Write the final O to gmem	COMMENT
LOW	…tn/third_party/flash-attention/hopper/epilogue_fwd.hpp	258	// Step 1: Write O from rmem -> smem	COMMENT
LOW	…tn/third_party/flash-attention/hopper/epilogue_fwd.hpp	288	// Step 2: Write LSE from rmem -> gmem	COMMENT
LOW	…tn/third_party/flash-attention/hopper/epilogue_fwd.hpp	317	// Step 3: Write O from smem -> gmem	COMMENT
LOW⚡	…s-paged-attn/src/cuda/backend/context_attention_mla.rs	147	// Step 1: Absorb w_uk into query	COMMENT
LOW⚡	…s-paged-attn/src/cuda/backend/context_attention_mla.rs	155	// Step 2: Gather cached ckv/kpe if any context exists	COMMENT
LOW⚡	…s-paged-attn/src/cuda/backend/context_attention_mla.rs	165	// Step 3: Per-sequence attention in latent space	COMMENT
LOW	docs/src/content/docs/examples/python/tool-call.md	62	# Step 1: Model generates a tool call	COMMENT
LOW	docs/src/content/docs/examples/python/tool-call.md	78	# Step 2: Send the result back	COMMENT
LOW	docs/src/content/docs/examples/python/tool-call.md	87	# Step 3: Model produces the final answer	COMMENT
LOW⚡	docs/src/content/docs/examples/server/tool-calling.md	61	# Step 1: Model generates a tool call	COMMENT
LOW⚡	docs/src/content/docs/examples/server/tool-calling.md	68	# Step 2: Execute the tool locally	COMMENT
LOW⚡	docs/src/content/docs/examples/server/tool-calling.md	74	# Step 3: Send the result back	COMMENT
LOW	docs/src/content/docs/examples/server/tool-calling.md	100	# Step 4: Model produces the final answer	COMMENT
LOW	…content/docs/examples/server/anthropic-tool-calling.md	73	# Step 1: Model generates a tool call.	COMMENT
LOW	…content/docs/examples/server/anthropic-tool-calling.md	92	# Step 2: Execute the tool locally.	COMMENT
LOW	…content/docs/examples/server/anthropic-tool-calling.md	96	# Step 3: Send the result back.	COMMENT
LOW	…content/docs/examples/server/anthropic-tool-calling.md	111	# Step 4: Model produces the final answer.	COMMENT
LOW⚡	mistralrs-quant/src/gguf/cuda.rs	1498	// Step 1: Quantize input to Q8_1 (shared between gate and up)	COMMENT
LOW⚡	mistralrs-quant/src/gguf/cuda.rs	1503	// Step 2: Fused gate+up+activation+multiply	COMMENT
LOW⚡	mistralrs-quant/src/gguf/cuda.rs	1563	// Step 3: Quantize the intermediate output for down projection	COMMENT
LOW⚡	mistralrs-quant/src/gguf/cuda.rs	1573	// Step 4: Fused down+aggregate with topk_weights and atomicAdd	COMMENT
LOW	examples/python/tool_call.py	47	# Step 1: Model generates a tool call	COMMENT
LOW⚡	examples/python/tool_call.py	63	# Step 2: Send the result back	COMMENT
LOW⚡	examples/python/tool_call.py	72	# Step 3: Model produces the final answer	COMMENT
LOW	examples/server/anthropic_tool_calling.py	55	# Step 1: Model generates a tool call.	COMMENT
LOW⚡	examples/server/anthropic_tool_calling.py	74	# Step 2: Execute the tool locally.	COMMENT
LOW⚡	examples/server/anthropic_tool_calling.py	78	# Step 3: Send the result back.	COMMENT
LOW	examples/server/anthropic_tool_calling.py	93	# Step 4: Model produces the final answer.	COMMENT
LOW⚡	examples/server/tool_calling.py	43	# Step 1: Model generates a tool call	COMMENT
LOW⚡	examples/server/tool_calling.py	50	# Step 2: Execute the tool locally	COMMENT
LOW⚡	examples/server/tool_calling.py	56	# Step 3: Send the result back	COMMENT
LOW	examples/server/tool_calling.py	82	# Step 4: Model produces the final answer	COMMENT
LOW	mistralrs-core/src/utils/mod.rs	154	// Step 1: Add all choices to groups	COMMENT
LOW	mistralrs-core/src/utils/mod.rs	189	// Step 2: Respond with all groups	COMMENT
LOW	mistralrs-core/src/utils/mod.rs	252	// Step 3: Set state - This cannot be done in Step 2 as `group` is locking the refcell	COMMENT

Empty Error Handlers14 hits · 70 pts

Severity	File	Line	Snippet	Context
HIGH	releases/v0.9.3/scripts/bench_decode_serving.py	1150	except subprocess.TimeoutExpired:	CODE
HIGH	mistralrs-code-exec/python/executor.py	119	except Exception:	CODE
HIGH	mistralrs-code-exec/python/executor.py	130	except Exception:	CODE
HIGH	mistralrs-code-exec/python/executor.py	154	except Exception:	CODE
HIGH	mistralrs-code-exec/python/executor.py	173	except ImportError:	CODE
HIGH	mistralrs-code-exec/python/executor.py	175	except Exception:	CODE
HIGH	mistralrs-code-exec/python/executor.py	187	except ImportError:	CODE
HIGH	scripts/generate_uqff_card.py	77	except KeyboardInterrupt:	CODE
HIGH	scripts/build_wheels.py	154	except (subprocess.SubprocessError, FileNotFoundError, OSError):	CODE
HIGH	scripts/build_wheels.py	125	except (subprocess.TimeoutExpired, FileNotFoundError):	CODE
HIGH	scripts/build_wheels.py	197	except (subprocess.SubprocessError, FileNotFoundError, OSError):	CODE
HIGH	scripts/production_soak.py	1285	except asyncio.CancelledError:	CODE
HIGH	scripts/production_soak.py	8935	except asyncio.TimeoutError:	CODE
HIGH	scripts/production_soak.py	1265	except (asyncio.CancelledError, Exception):	CODE

Excessive Try-Catch Wrapping58 hits · 68 pts

Severity	File	Line	Snippet	Context
LOW	releases/v0.9.3/scripts/bench_decode_serving.py	1489	except Exception as error: # noqa: BLE001	CODE
LOW	mistralrs-code-exec/python/executor.py	119	except Exception:	CODE
LOW	mistralrs-code-exec/python/executor.py	130	except Exception:	CODE
LOW	mistralrs-code-exec/python/executor.py	154	except Exception:	CODE
LOW	mistralrs-code-exec/python/executor.py	175	except Exception:	CODE
LOW	mistralrs-code-exec/python/executor.py	308	except Exception as e:	CODE
LOW	mistralrs-code-exec/python/executor.py	384	except Exception:	CODE
LOW	mistralrs-code-exec/python/executor.py	471	except Exception as e:	CODE
LOW	docs/scripts/render_pyi.py	158	except Exception:	CODE
LOW	docs/src/content/docs/examples/python/agentic-tools.md	56	except Exception as e:	CODE
LOW	…s/src/content/docs/examples/python/test-multi-model.md	108	except Exception as e:	CODE
LOW	…s/src/content/docs/examples/python/test-multi-model.md	157	except Exception as e:	CODE
LOW	…s/src/content/docs/examples/python/test-multi-model.md	210	except Exception as e:	CODE
LOW⚡	…s/src/content/docs/examples/python/test-multi-model.md	238	except Exception as e:	CODE
LOW⚡	…s/src/content/docs/examples/python/test-multi-model.md	246	except Exception as e:	CODE
LOW⚡	…s/src/content/docs/examples/python/test-multi-model.md	252	except Exception as e:	CODE
LOW	…s/src/content/docs/examples/python/custom-tool-call.md	34	except Exception:	CODE
LOW	docs/src/content/docs/examples/python/custom-search.md	33	except Exception:	CODE
LOW	docs/src/content/docs/examples/server/mcp-chat.md	134	except Exception as e:	CODE
MEDIUM	docs/src/content/docs/examples/server/mcp-chat.md	135	print(f"Error making request: {e}")	CODE
LOW	docs/src/content/docs/examples/server/tool-dispatch.md	164	except Exception as e:	CODE
MEDIUM	docs/src/content/docs/examples/server/tool-dispatch.md	165	print(f"Error: {e}")	CODE
LOW	…rc/content/docs/examples/server/agentic-tool-rounds.md	87	except Exception as e:	CODE
MEDIUM	…rc/content/docs/examples/server/agentic-tool-rounds.md	88	print(f"Error: {e}")	CODE
LOW	…content/docs/examples/server/streaming-tool-calling.md	93	except Exception as e:	CODE
LOW	…content/docs/examples/server/streaming-tool-calling.md	192	except Exception as e:	CODE
MEDIUM	…s/src/content/docs/examples/server/multi-model-chat.md	287	print("Error: Could not connect to server.")	CODE
LOW	examples/python/test_multi_model.py	89	except Exception as e:	CODE
LOW	examples/python/test_multi_model.py	138	except Exception as e:	CODE
LOW	examples/python/test_multi_model.py	191	except Exception as e:	CODE
LOW⚡	examples/python/test_multi_model.py	219	except Exception as e:	CODE
LOW⚡	examples/python/test_multi_model.py	227	except Exception as e:	CODE
LOW⚡	examples/python/test_multi_model.py	233	except Exception as e:	CODE
LOW	examples/python/custom_search.py	19	except Exception:	CODE
LOW	examples/python/agentic_tools.py	34	except Exception as e:	CODE
LOW	examples/python/custom_tool_call.py	20	except Exception:	CODE
LOW	examples/server/agentic_tool_rounds.py	60	except Exception as e:	CODE
MEDIUM	examples/server/agentic_tool_rounds.py	61	print(f"Error: {e}")	CODE
LOW	examples/server/tool_dispatch.py	142	except Exception as e:	CODE
MEDIUM	examples/server/tool_dispatch.py	143	print(f"Error: {e}")	CODE
LOW	examples/server/streaming_tool_calling.py	64	except Exception as e:	CODE
LOW	examples/server/streaming_tool_calling.py	163	except Exception as e:	CODE
MEDIUM	examples/server/multi_model_chat.py	265	print("Error: Could not connect to server.")	CODE
LOW	examples/server/mcp_chat.py	109	except Exception as e:	CODE
MEDIUM	examples/server/mcp_chat.py	110	print(f"Error making request: {e}")	CODE
LOW	scripts/convert_awq_marlin.py	218	except Exception as e:	CODE
MEDIUM	scripts/convert_awq_marlin.py	219	print(f"Error: {e}")	CODE
MEDIUM	scripts/upload_wheels.py	244	print(f"Error: Directory does not exist: {args.directory}", file=sys.stderr)	STRING
MEDIUM	scripts/build_wheels.py	515	print(f"Error: {pkg} cannot be built on this platform", file=sys.stderr)	STRING
LOW	scripts/production_soak.py	1287	except Exception as exc:	CODE
LOW	scripts/production_soak.py	1591	except Exception as exc:	CODE
LOW	scripts/production_soak.py	1609	except Exception as exc:	CODE
LOW	scripts/production_soak.py	2944	except Exception as exc:	CODE
LOW	scripts/production_soak.py	4294	except Exception as exc:	CODE
LOW	scripts/production_soak.py	11767	except Exception as exc:	CODE
LOW	scripts/production_soak.py	14087	except Exception as exc:	CODE
LOW	scripts/production_soak.py	14181	except Exception as exc:	CODE
LOW	scripts/bench_nvfp4_serving.py	240	except Exception as error:	CODE

Verbosity Indicators34 hits · 63 pts

Severity	File	Line	Snippet	Context
LOW	mistralrs-pyo3/src/lib.rs	1337	// found when rig is used as client. In this case, we need to check if	COMMENT
LOW	mistralrs-pyo3/src/lib.rs	2298	// found when rig is used as client. In this case, we need to check if	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	458	// Step 1: load LSE_partial from gmem -> smem	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	503	// Step 2: Load O_partial from gmem -> smem for split = 0, 1, ..., kStages - 2.	COMMENT
LOW⚡	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	564	// Step 3: load and transpose LSE_partial from smem -> rmem	COMMENT
LOW⚡	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	574	// Step 4: compute the final LSE along the split dimension	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	623	// Step 5: store final LSE back to gmem	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	646	// Step 6: read O_partial from gmem -> smem -> rmem and accumulate the final O	COMMENT
LOW	…arty/flash-attention/hopper/flash_fwd_combine_kernel.h	689	// Step 7: Write the final O to gmem	COMMENT
LOW	…tn/third_party/flash-attention/hopper/epilogue_fwd.hpp	258	// Step 1: Write O from rmem -> smem	COMMENT
LOW	…tn/third_party/flash-attention/hopper/epilogue_fwd.hpp	288	// Step 2: Write LSE from rmem -> gmem	COMMENT
LOW	…tn/third_party/flash-attention/hopper/epilogue_fwd.hpp	317	// Step 3: Write O from smem -> gmem	COMMENT
LOW⚡	…s-paged-attn/src/cuda/backend/context_attention_mla.rs	147	// Step 1: Absorb w_uk into query	COMMENT
LOW⚡	…s-paged-attn/src/cuda/backend/context_attention_mla.rs	155	// Step 2: Gather cached ckv/kpe if any context exists	COMMENT
LOW⚡	…s-paged-attn/src/cuda/backend/context_attention_mla.rs	165	// Step 3: Per-sequence attention in latent space	COMMENT
LOW	mistralrs-server-core/src/chat_completion.rs	708	// found when rig is used as client. In this case, we need to check if	COMMENT
LOW⚡	mistralrs-quant/src/gguf/cuda.rs	1498	// Step 1: Quantize input to Q8_1 (shared between gate and up)	COMMENT
LOW⚡	mistralrs-quant/src/gguf/cuda.rs	1503	// Step 2: Fused gate+up+activation+multiply	COMMENT
LOW⚡	mistralrs-quant/src/gguf/cuda.rs	1563	// Step 3: Quantize the intermediate output for down projection	COMMENT
LOW⚡	mistralrs-quant/src/gguf/cuda.rs	1573	// Step 4: Fused down+aggregate with topk_weights and atomicAdd	COMMENT
LOW	examples/python/tool_call.py	47	# Step 1: Model generates a tool call	COMMENT
LOW⚡	examples/python/tool_call.py	63	# Step 2: Send the result back	COMMENT
LOW⚡	examples/python/tool_call.py	72	# Step 3: Model produces the final answer	COMMENT
LOW	examples/server/anthropic_tool_calling.py	55	# Step 1: Model generates a tool call.	COMMENT
LOW⚡	examples/server/anthropic_tool_calling.py	74	# Step 2: Execute the tool locally.	COMMENT
LOW⚡	examples/server/anthropic_tool_calling.py	78	# Step 3: Send the result back.	COMMENT
LOW	examples/server/anthropic_tool_calling.py	93	# Step 4: Model produces the final answer.	COMMENT
LOW⚡	examples/server/tool_calling.py	43	# Step 1: Model generates a tool call	COMMENT
LOW⚡	examples/server/tool_calling.py	50	# Step 2: Execute the tool locally	COMMENT
LOW⚡	examples/server/tool_calling.py	56	# Step 3: Send the result back	COMMENT
LOW	examples/server/tool_calling.py	82	# Step 4: Model produces the final answer	COMMENT
LOW	mistralrs-core/src/utils/mod.rs	154	// Step 1: Add all choices to groups	COMMENT
LOW	mistralrs-core/src/utils/mod.rs	189	// Step 2: Respond with all groups	COMMENT
LOW	mistralrs-core/src/utils/mod.rs	252	// Step 3: Set state - This cannot be done in Step 2 as `group` is locking the refcell	COMMENT

AI Slop Vocabulary14 hits · 40 pts

Severity	File	Line	Snippet	Context
MEDIUM	mistralrs-mcp/src/client.rs	177	/// - Tool Registration: Converts MCP tools to internal Tool format for seamless integration	COMMENT
MEDIUM	mistralrs-quant/src/distributed/layers.rs	2650	// Keep weights as FP8 using BlockwiseFP8 to leverage native FP8 GEMM in gather_forward	COMMENT
MEDIUM	mistralrs-quant/src/distributed/layers.rs	2809	// Keep weights as FP8 using BlockwiseFP8 to leverage native FP8 GEMM in gather_forward	COMMENT
MEDIUM	scripts/generate_uqff_card.py	84	output += "\n\n## Topologies\n**The following model topologies were used to generate this UQFF file. Only informatio	STRING
MEDIUM	mistralrs-core/src/model_selected.rs	93	/// Generate and utilize an imatrix to enhance GGUF quantizations.	COMMENT
MEDIUM	mistralrs-core/src/model_selected.rs	182	/// Generate and utilize an imatrix to enhance GGUF quantizations.	COMMENT
MEDIUM	mistralrs-core/src/model_selected.rs	318	/// Generate and utilize an imatrix to enhance GGUF quantizations.	COMMENT
MEDIUM	mistralrs-core/src/model_selected.rs	727	/// Generate and utilize an imatrix to enhance GGUF quantizations.	COMMENT
MEDIUM	mistralrs-core/src/toml_selector.rs	75	/// Generate and utilize an imatrix to enhance GGUF quantizations.	COMMENT
MEDIUM	mistralrs-core/src/toml_selector.rs	465	/// Generate and utilize an imatrix to enhance GGUF quantizations.	COMMENT
MEDIUM	mistralrs-core/src/vision_models/siglip.rs	89	// For robust handling of NaNs, you might need a custom comparison.	COMMENT
MEDIUM	mistralrs-core/src/vision_models/idefics3/vision.rs	98	// For robust handling of NaNs, you might need a custom comparison.	COMMENT
MEDIUM	mistralrs-core/src/vision_models/idefics2/mod.rs	286	// For robust handling of NaNs, you might need a custom comparison.	COMMENT
MEDIUM	mistralrs/examples/advanced/perplexity/main.rs	32	/// Generate and utilize an imatrix to enhance GGUF quantizations.	COMMENT

Deep Nesting42 hits · 33 pts

Severity	File	Line	Context
LOW	releases/v0.9.0/scripts/plot_results.py	41	CODE
LOW	releases/v0.9.0/scripts/plot_results.py	67	CODE
LOW	releases/v0.9.0/scripts/plot_results.py	96	CODE
LOW	releases/v0.9.3/scripts/merge_results.py	74	CODE
LOW	releases/v0.9.3/scripts/merge_results.py	193	CODE
LOW	releases/v0.9.3/scripts/bench_decode_serving.py	1511	CODE
LOW	releases/v0.9.3/scripts/bench_decode_serving.py	1558	CODE
LOW	releases/v0.9.3/scripts/bench_decode_serving.py	1657	CODE
LOW	mistralrs-code-exec/python/executor.py	272	CODE
LOW	mistralrs-code-exec/python/executor.py	325	CODE
LOW	docs/scripts/render_pyi.py	255	CODE
LOW	docs/scripts/render_pyi.py	437	CODE
LOW	examples/python/custom_search.py	11	CODE
LOW	examples/python/custom_tool_call.py	12	CODE
LOW	examples/server/streaming_tool_calling.py	77	CODE
LOW	examples/server/multi_model_chat.py	57	CODE
LOW	examples/server/anthropic_streaming.py	19	CODE
LOW	examples/server/code_execution_approval.py	87	CODE
LOW	examples/server/anthropic_agentic.py	19	CODE
LOW	scripts/production_soak.py	502	CODE
LOW	scripts/production_soak.py	1376	CODE
LOW	scripts/production_soak.py	1454	CODE
LOW	scripts/production_soak.py	2045	CODE
LOW	scripts/production_soak.py	2683	CODE
LOW	scripts/production_soak.py	3141	CODE
LOW	scripts/production_soak.py	3862	CODE
LOW	scripts/production_soak.py	8792	CODE
LOW	scripts/production_soak.py	9021	CODE
LOW	scripts/production_soak.py	11058	CODE
LOW	scripts/production_soak.py	11308	CODE
LOW	scripts/production_soak.py	11388	CODE
LOW	scripts/production_soak.py	13292	CODE
LOW	scripts/production_soak.py	14070	CODE
LOW	scripts/production_soak.py	508	CODE
LOW	scripts/production_soak.py	595	CODE
LOW	scripts/production_soak.py	707	CODE
LOW	scripts/production_soak.py	1022	CODE
LOW	scripts/test_release_cuda_cache.py	44	CODE
LOW	scripts/bench_nvfp4_serving.py	62	CODE
LOW	scripts/bench_nvfp4_serving.py	164	CODE
LOW	scripts/tests/test_install.py	305	CODE
LOW	scripts/tests/test_install.py	401	CODE

Self-Referential Comments6 hits · 18 pts

Severity	File	Line	Snippet	Context
MEDIUM	examples/python/test_multi_model.py	26	# Create a simple runner	COMMENT
MEDIUM	examples/python/multi_model_example.py	29	# Create a runner with Gemma 4 E4B multimodal model	COMMENT
MEDIUM	examples/python/gpt_oss.py	11	# Create a GPT-OSS model runner	COMMENT
MEDIUM	examples/python/granite.py	8	# Create a Granite model runner	COMMENT
MEDIUM	examples/python/smollm3.py	8	# Create a SmolLM3 model runner	COMMENT
MEDIUM	examples/server/mcp_chat.py	49	# Create a chat completion request that can trigger MCP tool usage	COMMENT

Redundant / Tautological Comments7 hits · 10 pts

Severity	File	Line	Snippet	Context
LOW	install.sh	783	# Check if Rust is installed	COMMENT
LOW	install.sh	906	# Check if MKL is installed	COMMENT
LOW	install.sh	923	# Check if CPU is Intel	COMMENT
LOW	install.sh	973	# Check if cuDNN is installed	COMMENT
LOW	install.sh	984	# Check if NCCL is installed	COMMENT
LOW	install.sh	1054	# Check if ffmpeg is installed	COMMENT
LOW	examples/python/multi_model_example.py	81	# Check if a model is loaded	COMMENT

Unused Imports10 hits · 10 pts

Severity	File	Line	Context
LOW	releases/v0.9.0/scripts/bench_cpu_sweep.py	17	CODE
LOW	releases/v0.9.0/scripts/plot_results.py	5	CODE
LOW	releases/v0.9.3/scripts/merge_results.py	4	CODE
LOW	releases/v0.9.3/scripts/bench_decode_serving.py	4	CODE
LOW	docs/scripts/render_pyi.py	13	CODE
LOW	docs/scripts/render_examples.py	15	CODE
LOW	scripts/upload_wheels.py	14	CODE
LOW	scripts/build_wheels.py	14	CODE
LOW	scripts/production_soak.py	4	CODE
LOW	scripts/bench_nvfp4_serving.py	4	CODE

Fake / Example Data6 hits · 6 pts

Severity	File	Line	Snippet	Context
LOW	docs/src/content/docs/examples/server/mcp-chat.md	52	api_key="placeholder", # mistral.rs doesn't require a real API key	CODE
LOW	docs/src/content/docs/examples/server/tool-dispatch.md	119	api_key="placeholder",	CODE
LOW	…rc/content/docs/examples/server/agentic-tool-rounds.md	54	api_key="placeholder",	CODE
LOW	examples/server/agentic_tool_rounds.py	27	api_key="placeholder",	CODE
LOW	examples/server/tool_dispatch.py	97	api_key="placeholder",	CODE
LOW	examples/server/mcp_chat.py	27	api_key="placeholder", # mistral.rs doesn't require a real API key	CODE

Dead Code1 hit · 2 pts

Severity	File	Line	Snippet	Context
MEDIUM	scripts/test_production_soak.py	515		CODE

Modern Structural Boilerplate1 hit · 1 pts

Severity	File	Line	Snippet	Context
LOW	releases/v0.9.3/scripts/bench_decode_serving.py	1089	def update_manifest(manifest_path: Path, manifest: dict) -> None:	CODE

AI Structural Patterns1 hit · 0 pts

Severity	File	Line	Snippet	Context
LOW	scripts/test_production_soak.py	14		CODE

EricLBuehler/mistral.rs

Open-Source Code Quality, Security and Technical Debt Review