mcp-ai-agent-guidelines/dist/strategies/speckit/tasks-parser.js

A comprehensive Model Context Protocol server providing advanced tools, resources, and prompts for implementing AI agent best practices

/**
 * Tasks parser - extracts task data from tasks.md markdown
 *
 * @module strategies/speckit/tasks-parser
 */
/**
 * Parse tasks.md markdown content into structured Tasks
 *
 * Extracts task items from markdown format with task ID patterns like
 * TASK-001, P4-019, etc.
 *
 * @param markdown - The tasks.md markdown content to parse
 * @returns Structured Tasks object with items array
 *
 * @example
 * ```typescript
 * const markdown = `
 * # Tasks
 *
 * ## Phase 1
 *
 * ### TASK-001: Setup database
 * - **Priority**: high
 * - **Estimate**: 2 days
 * - **Description**: Initialize PostgreSQL database
 * - **Acceptance Criteria**:
 *   - Database is running
 *   - Schema is created
 * `;
 *
 * const tasks = parseTasksFromMarkdown(markdown);
 * // Returns: { items: [{ id: "TASK-001", title: "Setup database", ... }] }
 * ```
 */
export function parseTasksFromMarkdown(markdown) {
    const items = [];
    // Pattern to match task headings like "### TASK-001: Title" or "### P4-019: Title"
    const taskHeadingPattern = /^###\s+([A-Z0-9]+-\d+):\s*(.+)$/gm;
    let match;
    // biome-ignore lint/suspicious/noAssignInExpressions: Required for regex iteration
    while ((match = taskHeadingPattern.exec(markdown)) !== null) {
        const taskId = match[1];
        const title = match[2].trim();
        // Extract task details from the section following this heading
        const sectionStart = match.index + match[0].length;
        const nextHeadingMatch = markdown.slice(sectionStart).match(/^###?\s+/m);
        const sectionEnd = nextHeadingMatch
            ? sectionStart + (nextHeadingMatch.index ?? 0)
            : markdown.length;
        const taskSection = markdown.slice(sectionStart, sectionEnd);
        // Extract task properties from bullet points
        const task = {
            id: taskId,
            title,
            description: extractDescription(taskSection),
            priority: extractPriority(taskSection),
            estimate: extractEstimate(taskSection),
            phase: extractPhase(taskSection),
            acceptanceCriteria: extractAcceptanceCriteria(taskSection),
            dependencies: extractDependencies(taskSection),
        };
        items.push(task);
    }
    return {
        items,
        metadata: {
            createdAt: new Date().toISOString(),
            lastUpdated: new Date().toISOString(),
        },
    };
}
/**
 * Extract description from task section
 */
function extractDescription(section) {
    // Look for "- **Description**: ..." or "**Description**: ..."
    const descMatch = section.match(/^\s*[-*]?\s*\*\*Description\*\*:\s*(.+)$/im);
    if (descMatch) {
        return descMatch[1].trim();
    }
    // Fallback: use first paragraph after heading
    const lines = section.trim().split("\n");
    for (const line of lines) {
        const trimmed = line.trim();
        if (trimmed && !trimmed.startsWith("-") && !trimmed.startsWith("*")) {
            return trimmed;
        }
    }
    return "";
}
/**
 * Extract priority from task section
 */
function extractPriority(section) {
    const priorityMatch = section.match(/^\s*[-*]?\s*\*\*Priority\*\*:\s*(\w+)/im);
    if (priorityMatch) {
        const priority = priorityMatch[1].toLowerCase();
        if (priority === "high" || priority === "medium" || priority === "low") {
            return priority;
        }
    }
    return "medium"; // default
}
/**
 * Extract estimate from task section
 */
function extractEstimate(section) {
    const estimateMatch = section.match(/^\s*[-*]?\s*\*\*Estimate\*\*:\s*(.+)$/im);
    return estimateMatch ? estimateMatch[1].trim() : "";
}
/**
 * Extract phase from task section
 */
function extractPhase(section) {
    const phaseMatch = section.match(/^\s*[-*]?\s*\*\*Phase\*\*:\s*(.+)$/im);
    return phaseMatch ? phaseMatch[1].trim() : undefined;
}
/**
 * Extract acceptance criteria from task section
 */
function extractAcceptanceCriteria(section) {
    const criteria = [];
    // Look for "- **Acceptance Criteria**:" followed by nested bullets
    const criteriaMatch = section.match(/^\s*[-*]?\s*\*\*Acceptance Criteria\*\*:\s*$/im);
    if (criteriaMatch) {
        const startIdx = (criteriaMatch.index !== undefined ? criteriaMatch.index : 0) +
            criteriaMatch[0].length;
        const remainingText = section.slice(startIdx);
        // Extract nested bullet points (indented with 2+ spaces)
        const nestedBulletPattern = /^\s{2,}[-*]\s+(.+)$/gm;
        let bulletMatch;
        // biome-ignore lint/suspicious/noAssignInExpressions: Required for regex iteration
        while ((bulletMatch = nestedBulletPattern.exec(remainingText)) !== null) {
            criteria.push(bulletMatch[1].trim());
        }
    }
    return criteria;
}
/**
 * Extract dependencies from task section
 */
function extractDependencies(section) {
    const dependencies = [];
    // Look for "- **Dependencies**: TASK-001, TASK-002" or similar
    const depsMatch = section.match(/^\s*[-*]?\s*\*\*Dependencies\*\*:\s*(.+)$/im);
    if (depsMatch) {
        const depsText = depsMatch[1];
        // Extract task IDs from the text
        const taskIdPattern = /\b([A-Z0-9]+-\d+)\b/g;
        let idMatch;
        // biome-ignore lint/suspicious/noAssignInExpressions: Required for regex iteration
        while ((idMatch = taskIdPattern.exec(depsText)) !== null) {
            dependencies.push(idMatch[1]);
        }
    }
    return dependencies.length > 0 ? dependencies : undefined;
}
//# sourceMappingURL=tasks-parser.js.map