Struct Rule 

pub struct Rule {

    pub id: RuleId,
    pub language: Language,
    pub severity: Severity,
    pub message: String,
    pub note: Option<String>,
    pub category: Option<Category>,
    pub fix: Option<String>,
    pub rewrite: Option<String>,
    pub autofix_safe: bool,
    pub autofix_rule: Value,
    pub rule: Value,
    pub constraints: Value,
    pub metric: Option<Metric>,
    pub threshold: Option<f64>,
    pub min_lines: Option<usize>,
    pub cross_file: Option<CrossFileKind>,
    pub resolution: Option<ResolutionKind>,
    pub source_dir: Option<PathBuf>,
    pub origin: RuleOrigin,
    pub files: Option<Vec<String>>,
    pub ignores: Option<Vec<String>>,
    pub enabled: bool,
    pub skip_test_code: bool,
    pub tests: Option<RuleTests>,
    pub ai_check: Option<AiCheck>,
}

Fields

id: RuleId

language: Language

severity: Severity

message: String

note: Option<String>

category: Option<Category>

fix: Option<String>

rewrite: Option<String>

ast-grep rewrite (fix) template applied to this rule’s match by scan --fix. Distinct from fix, which is a free-text human message: rewrite carries the replacement pattern (it may reference the rule’s metavariables). The AST detection pass never reads this field.

autofix_safe: bool

Opt-in marker that this rule’s rewrite is safe to apply automatically (idempotent, no semantic change). scan --fix ignores the rewrite of any rule without it. Defaults to false. On a rule from custom_dirs or [[rules.sources]] it is not enough: the rule id must also be listed in the user config’s fix.allow_external.

autofix_rule: Value

Matcher used only by scan --fix to locate nodes to rewrite, when it must be narrower than the detection rule (detect broadly, fix narrowly). A rule can flag every variant of a pattern while auto-rewriting only the subset whose rewrite is provably safe. null (the default) means --fix reuses rule. The detection pass never reads this field. --fix only rewrites an autofix_rule match whose range lies inside a match of the detection rule (with its constraints), so it can narrow but never widen what detection reports.

rule: Value

constraints: Value

ast-grep metavariable constraints, a sibling of rule. Passed through verbatim so a rule can restrict a captured metavariable (e.g. bind the receiver of $R.clone() to an already-owned value). null when unset.

metric: Option<Metric>

File-level metric evaluated instead of rule. Mutually exclusive with it.

threshold: Option<f64>

The value metric must exceed for the rule to fire. Required with metric.

min_lines: Option<usize>

Minimum file length in lines before a metric rule is evaluated. Below it the file is skipped, so ratios are not distorted by tiny files (a small module full of required doc comments). None means no floor.

cross_file: Option<CrossFileKind>

Project-wide analysis evaluated instead of rule or metric, and mutually exclusive with both. The kind selects builtin Rust logic; the YAML only carries the rule’s identity, severity and user-facing text.

resolution: Option<ResolutionKind>

Per-file import resolution evaluated instead of rule, metric or cross_file, and mutually exclusive with all three. The kind selects builtin resolution logic; the YAML only carries the rule’s identity, severity and user-facing text. $import in message is replaced with the unresolved specifier.

source_dir: Option<PathBuf>

Directory a custom rule was loaded from, used to resolve test fixture paths. None for builtin rules, which resolve against BuiltinRules.

origin: RuleOrigin

Provenance of this rule (builtin, git source, or local path), stamped at load time and reported by slopguard list. Never present in YAML.

files: Option<Vec<String>>

ignores: Option<Vec<String>>

enabled: bool

Drop findings located inside #[cfg(test)] blocks of scanned Rust files. Complements ignores, which only filters whole files by path.

skip_test_code: bool

tests: Option<RuleTests>

ai_check: Option<AiCheck>

When set, matches of rule are candidates confirmed by an LLM before being reported. See AiCheck. The AST scanner ignores this field.

Implementations

impl Rule

pub fn effective_category(&self) -> Category

The rule’s category, resolving an unset category to the default (Correctness), the way the scanner treats it.

pub fn is_metric(&self) -> bool

Whether this rule measures a file-level metric instead of matching AST.

pub fn metric_spec(&self) -> Option<(Metric, f64)>

The metric and threshold of a file-level rule, None for AST rules.

pub fn is_cross_file(&self) -> bool

Whether this rule is evaluated by the project-wide pass instead of per file.

pub fn is_resolution(&self) -> bool

Whether this rule is evaluated by the per-file import resolution pass.

pub fn is_autofixable(&self) -> bool

Whether scan --fix may rewrite this rule’s matches: it carries a rewrite template and is explicitly marked autofix_safe. An external (non-builtin) rule additionally needs its id in fix.allow_external, which scan --fix checks on top of this.

pub fn fix_matcher(&self) -> &Value

The matcher scan --fix uses to find nodes to rewrite: the dedicated autofix_rule when set, otherwise the detection rule.

pub fn cross_file_kind(&self) -> Option<CrossFileKind>

The cross-file analysis this rule requests, None for AST and metric rules.

pub fn resolution_kind(&self) -> Option<ResolutionKind>

The resolution analysis this rule requests, None for all other rules.

Trait Implementations

impl Clone for Rule

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Rule

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for Rule

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for Rule

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.