reasonkit/thinktool/

yaml_loader.rs

//! YAML Protocol Loader
//!
//! Loads ThinkTool protocols from YAML files (thinktools.yaml format).

use serde::{Deserialize, Serialize};
use std::collections::HashMap;
use std::path::Path;

use super::profiles::ReasoningProfile;
use super::protocol::{
    AggregationType, CritiqueSeverity, DecisionMethod, InputSpec, OutputSpec, Protocol,
    ProtocolMetadata, ProtocolStep, ReasoningStrategy, StepAction, StepOutputFormat,
};
use crate::error::{Error, Result};

/// YAML ThinkTool module definition
#[derive(Debug, Clone, Deserialize, Serialize)]
struct YamlThinkToolModule {
    id: String,
    name: String,
    #[serde(default)]
    shortcode: String,
    category: String,
    tier: String,
    description: String,
    capabilities: Vec<String>,
    #[serde(default)]
    output_schema: String,
    parameters: HashMap<String, serde_json::Value>,
    #[serde(default)]
    confidence_factors: Vec<YamlConfidenceFactor>,
    thinking_pattern: YamlThinkingPattern,
    #[serde(default)]
    typical_duration: String,
    #[serde(default)]
    token_cost_estimate: String,
}

#[derive(Debug, Clone, Deserialize, Serialize)]
struct YamlConfidenceFactor {
    factor: String,
    weight: f64,
    formula: String,
}

#[derive(Debug, Clone, Deserialize, Serialize)]
struct YamlThinkingPattern {
    #[serde(rename = "type")]
    pattern_type: String,
    steps: Vec<String>,
}

/// Root YAML structure for thinktools.yaml
#[derive(Debug, Clone, Deserialize, Serialize)]
struct YamlThinkToolsV2 {
    version: String,
    schema: String,
    #[serde(default)]
    thinktool_modules: HashMap<String, YamlThinkToolModule>,
}

/// Load protocols from the thinktools.yaml file
pub fn load_from_yaml_file(path: &Path) -> Result<Vec<Protocol>> {
    let content = std::fs::read_to_string(path).map_err(|e| Error::IoMessage {
        message: format!("Failed to read YAML file {}: {}", path.display(), e),
    })?;

    load_from_yaml_string(&content)
}

/// Load protocols from a YAML string
pub fn load_from_yaml_string(yaml_content: &str) -> Result<Vec<Protocol>> {
    let yaml_data: YamlThinkToolsV2 =
        serde_yaml::from_str(yaml_content).map_err(|e| Error::Parse {
            message: format!("Failed to parse YAML: {}", e),
        })?;

    let mut protocols = Vec::new();

    for (module_key, module) in yaml_data.thinktool_modules {
        let protocol = convert_yaml_module_to_protocol(&module_key, &module)?;
        protocols.push(protocol);
    }

    Ok(protocols)
}

/// Load profiles from a YAML file
pub fn load_profiles_from_yaml_file(path: &Path) -> Result<Vec<ReasoningProfile>> {
    let content = std::fs::read_to_string(path).map_err(|e| Error::IoMessage {
        message: format!("Failed to read YAML file {}: {}", path.display(), e),
    })?;

    let profiles: Vec<ReasoningProfile> =
        serde_yaml::from_str(&content).map_err(|e| Error::Parse {
            message: format!("Failed to parse profiles YAML: {}", e),
        })?;

    Ok(profiles)
}

/// Convert a YAML module definition to a Protocol struct
fn convert_yaml_module_to_protocol(
    module_key: &str,
    yaml_module: &YamlThinkToolModule,
) -> Result<Protocol> {
    // Determine reasoning strategy from category
    let strategy = match yaml_module.category.as_str() {
        "divergent" => ReasoningStrategy::Expansive,
        "convergent" => ReasoningStrategy::Deductive,
        "foundational" => ReasoningStrategy::Analytical,
        "verification" => ReasoningStrategy::Verification,
        "adversarial" => ReasoningStrategy::Adversarial,
        _ => ReasoningStrategy::Analytical,
    };

    // Build input spec based on module type
    let input = build_input_spec(module_key);

    // Build steps from thinking pattern
    let steps = build_steps_from_pattern(&yaml_module.thinking_pattern, module_key)?;

    // Build output spec
    let output = build_output_spec(&yaml_module.name);

    // Build metadata
    let metadata = ProtocolMetadata {
        category: yaml_module.category.clone(),
        composable_with: get_composable_modules(module_key),
        typical_tokens: estimate_tokens(&yaml_module.token_cost_estimate),
        estimated_latency_ms: estimate_latency(&yaml_module.typical_duration),
        ..Default::default()
    };

    let protocol = Protocol {
        id: module_key.to_string(),
        name: yaml_module.name.clone(),
        version: "2.0.0".to_string(),
        description: yaml_module.description.trim().to_string(),
        strategy,
        input,
        steps,
        output,
        validation: Vec::new(),
        metadata,
    };

    // Validate the protocol
    protocol.validate().map_err(|errors| {
        Error::Validation(format!(
            "Invalid protocol {}: {}",
            protocol.id,
            errors.join(", ")
        ))
    })?;

    Ok(protocol)
}

/// Build input specification based on module type
fn build_input_spec(module_key: &str) -> InputSpec {
    match module_key {
        "gigathink" => InputSpec {
            required: vec!["query".to_string()],
            optional: vec!["context".to_string(), "constraints".to_string()],
        },
        "laserlogic" => InputSpec {
            required: vec!["argument".to_string()],
            optional: vec!["context".to_string()],
        },
        "bedrock" => InputSpec {
            required: vec!["statement".to_string()],
            optional: vec!["domain".to_string()],
        },
        "proofguard" => InputSpec {
            required: vec!["claim".to_string()],
            optional: vec!["sources".to_string()],
        },
        "brutalhonesty" => InputSpec {
            required: vec!["work".to_string()],
            optional: vec!["criteria".to_string()],
        },
        _ => InputSpec::default(),
    }
}

/// Build protocol steps from thinking pattern
fn build_steps_from_pattern(
    _pattern: &YamlThinkingPattern,
    module_key: &str,
) -> Result<Vec<ProtocolStep>> {
    match module_key {
        "gigathink" => Ok(build_gigathink_steps()),
        "laserlogic" => Ok(build_laserlogic_steps()),
        "bedrock" => Ok(build_bedrock_steps()),
        "proofguard" => Ok(build_proofguard_steps()),
        "brutalhonesty" => Ok(build_brutalhonesty_steps()),
        "powercombo" => Ok(build_gigathink_steps()), // Fallback to GigaThink for now
        _ => Err(Error::Validation(format!(
            "Unknown module type: {}",
            module_key
        ))),
    }
}

/// Build GigaThink steps
fn build_gigathink_steps() -> Vec<ProtocolStep> {
    vec![
        ProtocolStep {
            id: "identify_dimensions".to_string(),
            action: StepAction::Generate {
                min_count: 5,
                max_count: 10,
            },
            prompt_template:
                r#"Identify 5-10 distinct dimensions or angles to analyze this question:

Question: {{query}}
{{#if context}}Context: {{context}}{{/if}}
{{#if constraints}}Constraints: {{constraints}}{{/if}}

For each dimension, provide a brief label. Format as a numbered list."#
                    .to_string(),
            output_format: StepOutputFormat::List,
            min_confidence: 0.7,
            depends_on: vec![],
            branch: None,
        },
        ProtocolStep {
            id: "explore_perspectives".to_string(),
            action: StepAction::Analyze {
                criteria: vec![
                    "novelty".to_string(),
                    "relevance".to_string(),
                    "depth".to_string(),
                ],
            },
            prompt_template: r#"For each dimension identified, provide:
1. Key insight from this perspective
2. Supporting evidence or example
3. Implications or consequences
4. Confidence score (0.0-1.0)

Dimensions to explore:
{{identify_dimensions}}

Question: {{query}}"#
                .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.6,
            depends_on: vec!["identify_dimensions".to_string()],
            branch: None,
        },
        ProtocolStep {
            id: "synthesize".to_string(),
            action: StepAction::Synthesize {
                aggregation: AggregationType::ThematicClustering,
            },
            prompt_template:
                r#"Synthesize the perspectives into key themes and actionable insights:

Perspectives:
{{explore_perspectives}}

Provide:
1. Major themes (2-4)
2. Key insights (3-5)
3. Recommended actions (if applicable)
4. Areas of uncertainty"#
                    .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.8,
            depends_on: vec!["explore_perspectives".to_string()],
            branch: None,
        },
    ]
}

/// Build LaserLogic steps
fn build_laserlogic_steps() -> Vec<ProtocolStep> {
    vec![
        ProtocolStep {
            id: "extract_claims".to_string(),
            action: StepAction::Analyze {
                criteria: vec!["clarity".to_string(), "completeness".to_string()],
            },
            prompt_template: r#"Extract the logical structure from this argument:

Argument: {{argument}}

Identify:
1. Main conclusion
2. Supporting premises
3. Implicit assumptions
4. Causal claims (if any)

Format each as a clear statement."#
                .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.7,
            depends_on: vec![],
            branch: None,
        },
        ProtocolStep {
            id: "check_validity".to_string(),
            action: StepAction::Validate {
                rules: vec![
                    "logical_consistency".to_string(),
                    "premise_support".to_string(),
                ],
            },
            prompt_template: r#"Evaluate the logical validity of this argument analysis:

{{extract_claims}}

Based on the claims identified above, check:
1. Do the premises logically lead to the conclusion?
2. Are there gaps in the reasoning chain?
3. Is the argument valid (logical structure) vs sound (true premises)?
4. Rate the logical strength (0.0-1.0) with justification"#
                .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.8,
            depends_on: vec!["extract_claims".to_string()],
            branch: None,
        },
        ProtocolStep {
            id: "detect_fallacies".to_string(),
            action: StepAction::Critique {
                severity: CritiqueSeverity::Standard,
            },
            prompt_template: r#"Check for logical fallacies in the argument:

Argument structure:
{{extract_claims}}

Common fallacies to check:
- Ad hominem, Straw man, False dichotomy
- Appeal to authority, Circular reasoning
- Hasty generalization, Post hoc
- Slippery slope, Red herring

For each fallacy found, explain where and why."#
                .to_string(),
            output_format: StepOutputFormat::List,
            min_confidence: 0.7,
            depends_on: vec!["extract_claims".to_string()],
            branch: None,
        },
    ]
}

/// Build BedRock steps
fn build_bedrock_steps() -> Vec<ProtocolStep> {
    vec![
        ProtocolStep {
            id: "decompose".to_string(),
            action: StepAction::Analyze {
                criteria: vec!["fundamentality".to_string(), "independence".to_string()],
            },
            prompt_template: r#"Decompose this statement to first principles:

Statement: {{statement}}
{{#if domain}}Domain: {{domain}}{{/if}}

Ask repeatedly: "What is this based on? Why is this true?"
Continue until reaching fundamental axioms or assumptions.

Format as a tree structure showing dependencies."#
                .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.7,
            depends_on: vec![],
            branch: None,
        },
        ProtocolStep {
            id: "identify_axioms".to_string(),
            action: StepAction::Generate {
                min_count: 3,
                max_count: 7,
            },
            prompt_template: r#"From the decomposition, identify the foundational axioms:

Decomposition:
{{decompose}}

For each axiom:
1. State clearly
2. Explain why it's fundamental (cannot be further reduced)
3. Note if it's empirical, logical, or definitional
4. Rate certainty (0.0-1.0)"#
                .to_string(),
            output_format: StepOutputFormat::List,
            min_confidence: 0.8,
            depends_on: vec!["decompose".to_string()],
            branch: None,
        },
        ProtocolStep {
            id: "reconstruct".to_string(),
            action: StepAction::Synthesize {
                aggregation: AggregationType::WeightedMerge,
            },
            prompt_template: r#"Reconstruct the original statement from axioms:

Axioms:
{{identify_axioms}}

Original statement: {{statement}}

Show the logical path from axioms to statement.
Identify any gaps or leaps in reasoning.
Calculate overall confidence based on axiom certainties."#
                .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.75,
            depends_on: vec!["identify_axioms".to_string()],
            branch: None,
        },
    ]
}

/// Build ProofGuard steps
fn build_proofguard_steps() -> Vec<ProtocolStep> {
    vec![
        ProtocolStep {
            id: "identify_sources".to_string(),
            action: StepAction::CrossReference { min_sources: 3 },
            prompt_template: r#"Identify potential sources to verify this claim:

Claim: {{claim}}
{{#if sources}}Known sources: {{sources}}{{/if}}

List 3+ independent sources that could verify or refute this claim.
Prioritize: official docs, peer-reviewed, primary sources."#
                .to_string(),
            output_format: StepOutputFormat::List,
            min_confidence: 0.6,
            depends_on: vec![],
            branch: None,
        },
        ProtocolStep {
            id: "verify_each".to_string(),
            action: StepAction::Validate {
                rules: vec![
                    "source_reliability".to_string(),
                    "claim_support".to_string(),
                ],
            },
            prompt_template: r#"For each source, evaluate support for the claim:

Claim: {{claim}}
Sources to check:
{{identify_sources}}

For each source:
1. What does it say about the claim?
2. Support level: Confirms / Partially confirms / Neutral / Contradicts
3. Source reliability (0.0-1.0)
4. Key quote or evidence"#
                .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.7,
            depends_on: vec!["identify_sources".to_string()],
            branch: None,
        },
        ProtocolStep {
            id: "triangulate".to_string(),
            action: StepAction::Synthesize {
                aggregation: AggregationType::Consensus,
            },
            prompt_template: r#"Apply triangulation to determine claim validity:

Claim: {{claim}}
Source evaluations:
{{verify_each}}

Triangulation rules:
- 3+ independent confirming sources = HIGH confidence
- 2 confirming, 1 neutral = MEDIUM confidence
- Mixed results = LOW confidence, note discrepancies
- Any contradiction = FLAG for review

Provide final verdict and confidence score."#
                .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.8,
            depends_on: vec!["verify_each".to_string()],
            branch: None,
        },
    ]
}

/// Build BrutalHonesty steps
fn build_brutalhonesty_steps() -> Vec<ProtocolStep> {
    vec![
        ProtocolStep {
            id: "steelman".to_string(),
            action: StepAction::Analyze {
                criteria: vec!["strengths".to_string()],
            },
            prompt_template: r#"First, steelman the work - what are its genuine strengths?

Work to critique:
{{work}}

Identify:
1. What does this do well?
2. What problems does it solve?
3. What is genuinely valuable here?

Be generous but honest."#
                .to_string(),
            output_format: StepOutputFormat::List,
            min_confidence: 0.7,
            depends_on: vec![],
            branch: None,
        },
        ProtocolStep {
            id: "attack".to_string(),
            action: StepAction::Critique {
                severity: CritiqueSeverity::Brutal,
            },
            prompt_template: r#"Now be brutally honest - what's wrong with this?

Work:
{{work}}

Strengths identified:
{{steelman}}

Attack from all angles:
1. Logical flaws
2. Missing considerations
3. Weak assumptions
4. Implementation problems
5. Unintended consequences
6. What would a harsh critic say?

Don't hold back. Be specific."#
                .to_string(),
            output_format: StepOutputFormat::List,
            min_confidence: 0.6,
            depends_on: vec!["steelman".to_string()],
            branch: None,
        },
        ProtocolStep {
            id: "verdict".to_string(),
            action: StepAction::Decide {
                method: DecisionMethod::ProsCons,
            },
            prompt_template: r#"Final verdict - is this work acceptable?

Strengths:
{{steelman}}

Flaws:
{{attack}}

Provide:
1. Overall assessment (Pass / Conditional Pass / Fail)
2. Most critical issue to fix
3. Confidence in verdict (0.0-1.0)
4. What would make this excellent?"#
                .to_string(),
            output_format: StepOutputFormat::Structured,
            min_confidence: 0.75,
            depends_on: vec!["steelman".to_string(), "attack".to_string()],
            branch: None,
        },
    ]
}

/// Build output specification
fn build_output_spec(module_name: &str) -> OutputSpec {
    let format = format!("{}Result", module_name.replace(" ", ""));
    let fields = match module_name {
        "GigaThink" => vec![
            "dimensions".to_string(),
            "perspectives".to_string(),
            "themes".to_string(),
            "insights".to_string(),
            "confidence".to_string(),
        ],
        "LaserLogic" => vec![
            "conclusion".to_string(),
            "premises".to_string(),
            "validity".to_string(),
            "fallacies".to_string(),
            "confidence".to_string(),
        ],
        "BedRock" => vec![
            "axioms".to_string(),
            "decomposition".to_string(),
            "reconstruction".to_string(),
            "gaps".to_string(),
            "confidence".to_string(),
        ],
        "ProofGuard" => vec![
            "verdict".to_string(),
            "sources".to_string(),
            "evidence".to_string(),
            "discrepancies".to_string(),
            "confidence".to_string(),
        ],
        "BrutalHonesty" => vec![
            "strengths".to_string(),
            "flaws".to_string(),
            "verdict".to_string(),
            "critical_fix".to_string(),
            "confidence".to_string(),
        ],
        _ => vec!["confidence".to_string()],
    };

    OutputSpec { format, fields }
}

/// Get composable modules for a given module
fn get_composable_modules(module_key: &str) -> Vec<String> {
    match module_key {
        "gigathink" => vec!["laserlogic".to_string(), "brutalhonesty".to_string()],
        "laserlogic" => vec!["gigathink".to_string(), "bedrock".to_string()],
        "bedrock" => vec!["laserlogic".to_string(), "proofguard".to_string()],
        "proofguard" => vec!["bedrock".to_string(), "brutalhonesty".to_string()],
        "brutalhonesty" => vec!["gigathink".to_string(), "proofguard".to_string()],
        _ => vec![],
    }
}

/// Estimate token usage from cost estimate string
fn estimate_tokens(cost_estimate: &str) -> u32 {
    match cost_estimate {
        "low" => 1000,
        "medium" => 2000,
        "medium-high" => 2500,
        "high" => 3000,
        _ => 2000,
    }
}

/// Estimate latency from duration string
fn estimate_latency(duration: &str) -> u32 {
    // Parse duration strings like "30-90s" or "60-180s"
    if let Some(range) = duration.strip_suffix('s') {
        if let Some((low, high)) = range.split_once('-') {
            if let (Ok(low_val), Ok(high_val)) = (low.parse::<u32>(), high.parse::<u32>()) {
                return ((low_val + high_val) / 2) * 1000; // Convert to ms
            }
        }
    }
    5000 // Default 5 seconds
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_estimate_tokens() {
        assert_eq!(estimate_tokens("low"), 1000);
        assert_eq!(estimate_tokens("medium"), 2000);
        assert_eq!(estimate_tokens("high"), 3000);
    }

    #[test]
    fn test_estimate_latency() {
        assert_eq!(estimate_latency("30-90s"), 60000);
        assert_eq!(estimate_latency("60-180s"), 120000);
    }

    #[test]
    fn test_build_input_spec() {
        let spec = build_input_spec("gigathink");
        assert_eq!(spec.required, vec!["query"]);
        assert!(spec.optional.contains(&"context".to_string()));
    }
}