Struct PromptEvidence 

pub struct PromptEvidence {

    pub user_goal: Option<String>,
    pub project_summary: Option<String>,
    pub node_goal: Option<String>,
    pub output_files: Vec<String>,
    pub context_files: Vec<String>,
    pub verifier_diagnostics: Option<String>,
    pub previous_attempts: Vec<CorrectionAttemptRecord>,
    pub previous_attempt_count: usize,
    pub plugin_correction_fragment: Option<String>,
    pub legal_support_files: Vec<String>,
    pub existing_file_contents: Vec<(String, String)>,
    pub dependency_expectations: Option<DependencyExpectation>,
    pub rejected_bundle_summary: Option<String>,
    pub solo_file_path: Option<String>,
    pub solo_language: Option<String>,
    pub working_dir: Option<String>,
    pub active_plugins: Vec<String>,
    pub interface_signature: Option<String>,
    pub invariants: Option<String>,
    pub forbidden_patterns: Option<String>,
    pub weighted_tests: Option<String>,
    pub workspace_import_hints: Option<String>,
    pub evidence_section: Option<String>,
    pub error_feedback: Option<String>,
    pub restriction_map_context: Option<String>,
    pub project_file_tree: Option<String>,
    pub build_test_output: Option<String>,
    pub owner_plugin: Option<String>,
    pub energy_v_syn: Option<f32>,
}

Evidence inputs for the prompt compiler (PSP-7 §5).

Each prompt family draws from a different subset of these fields. The compiler ignores fields that are irrelevant for the selected intent.

Fields

user_goal: Option<String>

The user’s high-level goal or request.

project_summary: Option<String>

Project structure summary (file tree, detected languages, etc.).

node_goal: Option<String>

Node-scoped goal for actuator/verifier prompts.

output_files: Vec<String>

Output files this node is expected to produce.

context_files: Vec<String>

Context files the node should read.

verifier_diagnostics: Option<String>

Verifier diagnostics from the last verification pass.

previous_attempts: Vec<CorrectionAttemptRecord>

Previous correction attempt records for retry prompts.

previous_attempt_count: usize

Number of previous correction attempts (used when full records are unavailable).

plugin_correction_fragment: Option<String>

Plugin-contributed correction guidance.

legal_support_files: Vec<String>

Legal support files declared by the plugin.

existing_file_contents: Vec<(String, String)>

Existing file contents for context injection.

dependency_expectations: Option<DependencyExpectation>

Dependency expectations for the current task.

rejected_bundle_summary: Option<String>

Bundle that was rejected (for retarget prompts).

solo_file_path: Option<String>

Solo mode file path.

solo_language: Option<String>

Solo mode language hint.

working_dir: Option<String>

Working directory path for context-aware prompts.

active_plugins: Vec<String>

Active language plugins (e.g. ["rust", "python"]).

interface_signature: Option<String>

Contract interface signature for actuator/verifier prompts.

invariants: Option<String>

Contract invariants for actuator/verifier prompts.

forbidden_patterns: Option<String>

Contract forbidden patterns for actuator/verifier prompts.

weighted_tests: Option<String>

Contract weighted tests for verifier prompts.

workspace_import_hints: Option<String>

Workspace import hints for cross-module references.

evidence_section: Option<String>

Pre-formatted evidence section for architect prompts.

error_feedback: Option<String>

Error feedback from previous planning attempts.

restriction_map_context: Option<String>

Restriction map context for correction prompts (pre-formatted).

project_file_tree: Option<String>

Project file tree for correction prompts (pre-formatted lines).

build_test_output: Option<String>

Raw build/test output for correction prompts (truncated).

owner_plugin: Option<String>

Owner plugin name (e.g. “rust”, “python”) for language-specific guidance.

energy_v_syn: Option<f32>

Syntactic energy score from the last verification pass.

Trait Implementations

impl Clone for PromptEvidence

fn clone(&self) -> PromptEvidence

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PromptEvidence

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

Formats the value using the given formatter.

impl Default for PromptEvidence

fn default() -> PromptEvidence

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for PromptEvidence

fn deserialize<__D>( __deserializer: __D, ) -> Result<PromptEvidence, <__D as Deserializer<'de>>::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for PromptEvidence

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

where __S: Serializer,

Serialize this value into the given Serde serializer.