universal-agent-memory 0.1.0

package/dist/memory/prepopulate.js

@@ -0,0 +1,815 @@
+import { execSync } from 'child_process';
+import { existsSync, readFileSync, readdirSync, statSync } from 'fs';
+import { join, relative, extname, basename } from 'path';
+/**
+ * Prepopulate memory from project documentation, git history, and skills/artifacts
+ */
+export async function prepopulateMemory(cwd, options = {}) {
+    const shortTermMemories = [];
+    const longTermMemories = [];
+    const discoveredSkills = [];
+    const doAll = !options.docs && !options.git && !options.skills;
+    // Parse documentation
+    if (doAll || options.docs) {
+        const docs = await parseDocumentation(cwd, options.verbose);
+        const docMemories = extractDocumentationMemories(docs);
+        // Short-term: recent/important observations
+        shortTermMemories.push(...docMemories.filter(m => m.importance && m.importance >= 7).slice(0, 20));
+        // Long-term: all documentation memories
+        longTermMemories.push(...docMemories);
+    }
+    // Extract git learnings
+    if (doAll || options.git) {
+        const commits = await extractGitHistory(cwd, options.limit || 500, options.since);
+        const gitMemories = extractGitLearnings(commits);
+        // Short-term: recent significant commits
+        shortTermMemories.push(...gitMemories.filter(m => m.importance && m.importance >= 6).slice(0, 30));
+        // Long-term: all git learnings
+        longTermMemories.push(...gitMemories);
+    }
+    // Discover skills, droids, commands, and artifacts
+    if (doAll || options.skills) {
+        const skills = await discoverSkillsAndArtifacts(cwd, options.verbose);
+        discoveredSkills.push(...skills);
+        // Create memories for discovered skills
+        const skillMemories = extractSkillMemories(skills);
+        longTermMemories.push(...skillMemories);
+        // Add important skills to short-term
+        shortTermMemories.push(...skillMemories.filter(m => m.importance && m.importance >= 7).slice(0, 10));
+    }
+    return { shortTerm: shortTermMemories, longTerm: longTermMemories, skills: discoveredSkills };
+}
+/**
+ * Parse all documentation files in the project
+ */
+export async function parseDocumentation(cwd, verbose = false) {
+    const docs = [];
+    // Documentation file patterns to search for
+    const docPatterns = [
+        'README.md',
+        'README.txt',
+        'README',
+        'CONTRIBUTING.md',
+        'CHANGELOG.md',
+        'ARCHITECTURE.md',
+        'DEVELOPMENT.md',
+        'SECURITY.md',
+        '.github/PULL_REQUEST_TEMPLATE.md',
+        '.github/ISSUE_TEMPLATE.md',
+    ];
+    // Parse root-level docs
+    for (const pattern of docPatterns) {
+        const docPath = join(cwd, pattern);
+        if (existsSync(docPath)) {
+            const doc = parseMarkdownFile(docPath, cwd);
+            if (doc) {
+                docs.push(doc);
+                if (verbose)
+                    console.log(`  Parsed: ${pattern}`);
+            }
+        }
+    }
+    // Parse docs/ directory
+    const docsDir = join(cwd, 'docs');
+    if (existsSync(docsDir) && statSync(docsDir).isDirectory()) {
+        const docFiles = findMarkdownFiles(docsDir);
+        for (const docFile of docFiles) {
+            const doc = parseMarkdownFile(docFile, cwd);
+            if (doc) {
+                docs.push(doc);
+                if (verbose)
+                    console.log(`  Parsed: ${relative(cwd, docFile)}`);
+            }
+        }
+    }
+    // Parse ADRs (Architecture Decision Records)
+    const adrDirs = ['docs/adr', 'docs/decisions', 'adr', 'decisions'];
+    for (const adrDir of adrDirs) {
+        const adrPath = join(cwd, adrDir);
+        if (existsSync(adrPath) && statSync(adrPath).isDirectory()) {
+            const adrFiles = findMarkdownFiles(adrPath);
+            for (const adrFile of adrFiles) {
+                const doc = parseMarkdownFile(adrFile, cwd, 'adr');
+                if (doc) {
+                    docs.push(doc);
+                    if (verbose)
+                        console.log(`  Parsed ADR: ${relative(cwd, adrFile)}`);
+                }
+            }
+        }
+    }
+    return docs;
+}
+/**
+ * Extract git commit history with details
+ */
+export async function extractGitHistory(cwd, limit = 500, since) {
+    const commits = [];
+    try {
+        // Check if it's a git repo
+        execSync('git rev-parse --git-dir', { cwd, encoding: 'utf-8', stdio: 'pipe' });
+    }
+    catch {
+        return commits; // Not a git repo
+    }
+    // Build git log command
+    let gitCmd = `git log --pretty=format:"%H|%aI|%an|%s|%b<<<END>>>" -n ${limit}`;
+    if (since) {
+        gitCmd += ` --since="${since}"`;
+    }
+    try {
+        const output = execSync(gitCmd, { cwd, encoding: 'utf-8', maxBuffer: 50 * 1024 * 1024 });
+        const commitStrings = output.split('<<<END>>>').filter(s => s.trim());
+        for (const commitStr of commitStrings) {
+            const parts = commitStr.trim().split('|');
+            if (parts.length < 4)
+                continue;
+            const [hash, date, author, ...rest] = parts;
+            const messageAndBody = rest.join('|');
+            const [message, ...bodyParts] = messageAndBody.split('\n');
+            const body = bodyParts.join('\n').trim();
+            // Get files changed in this commit
+            let files = [];
+            try {
+                const filesOutput = execSync(`git diff-tree --no-commit-id --name-only -r ${hash}`, {
+                    cwd,
+                    encoding: 'utf-8',
+                    stdio: 'pipe',
+                });
+                files = filesOutput.trim().split('\n').filter(f => f);
+            }
+            catch {
+                // Ignore errors getting file list
+            }
+            // Determine commit type from conventional commit format
+            const type = categorizeCommit(message);
+            commits.push({
+                hash: hash.trim(),
+                date: date.trim(),
+                author: author.trim(),
+                message: message.trim(),
+                body,
+                files,
+                type,
+            });
+        }
+    }
+    catch (error) {
+        // Git command failed, return empty
+        console.warn('Failed to extract git history:', error);
+    }
+    return commits;
+}
+/**
+ * Convert documentation into memory entries
+ */
+export function extractDocumentationMemories(docs) {
+    const memories = [];
+    const now = new Date().toISOString();
+    for (const doc of docs) {
+        for (const section of doc.sections) {
+            // Skip empty or very short sections
+            if (!section.content || section.content.length < 50)
+                continue;
+            // Chunk large sections
+            const chunks = chunkContent(section.content, 500);
+            for (let i = 0; i < chunks.length; i++) {
+                const memoryType = mapSectionToMemoryType(section.type);
+                const importance = calculateDocImportance(section, doc.path);
+                memories.push({
+                    id: `doc-${basename(doc.path, extname(doc.path))}-${section.heading.toLowerCase().replace(/\s+/g, '-')}-${i}`,
+                    timestamp: now,
+                    type: memoryType,
+                    content: chunks[i],
+                    tags: extractTags(section, doc.path),
+                    importance,
+                    metadata: {
+                        source: 'documentation',
+                        file: doc.path,
+                        section: section.heading,
+                        docTitle: doc.title,
+                    },
+                });
+            }
+        }
+    }
+    return memories;
+}
+/**
+ * Convert git history into memory entries with learnings
+ */
+export function extractGitLearnings(commits) {
+    const memories = [];
+    // Group commits by type for analysis
+    const bugFixes = commits.filter(c => c.type === 'fix');
+    const features = commits.filter(c => c.type === 'feat');
+    const refactors = commits.filter(c => c.type === 'refactor');
+    const reverts = commits.filter(c => c.type === 'revert');
+    // Extract learnings from bug fixes
+    for (const commit of bugFixes.slice(0, 50)) {
+        const learning = extractBugFixLearning(commit);
+        if (learning) {
+            memories.push({
+                id: `git-fix-${commit.hash.substring(0, 8)}`,
+                timestamp: commit.date,
+                type: 'observation',
+                content: learning,
+                tags: ['bug-fix', ...extractFileTags(commit.files)],
+                importance: calculateCommitImportance(commit),
+                metadata: {
+                    source: 'git',
+                    commit: commit.hash,
+                    author: commit.author,
+                    files: commit.files.slice(0, 10),
+                },
+            });
+        }
+    }
+    // Extract learnings from features
+    for (const commit of features.slice(0, 30)) {
+        memories.push({
+            id: `git-feat-${commit.hash.substring(0, 8)}`,
+            timestamp: commit.date,
+            type: 'observation',
+            content: `Feature added: ${commit.message}${commit.body ? `. ${summarizeBody(commit.body)}` : ''}`,
+            tags: ['feature', ...extractFileTags(commit.files)],
+            importance: calculateCommitImportance(commit),
+            metadata: {
+                source: 'git',
+                commit: commit.hash,
+                author: commit.author,
+                files: commit.files.slice(0, 10),
+            },
+        });
+    }
+    // Extract learnings from refactors (architectural decisions)
+    for (const commit of refactors.slice(0, 20)) {
+        memories.push({
+            id: `git-refactor-${commit.hash.substring(0, 8)}`,
+            timestamp: commit.date,
+            type: 'thought',
+            content: `Architectural change: ${commit.message}${commit.body ? `. Reasoning: ${summarizeBody(commit.body)}` : ''}`,
+            tags: ['refactor', 'architecture', ...extractFileTags(commit.files)],
+            importance: calculateCommitImportance(commit) + 1, // Refactors are usually important decisions
+            metadata: {
+                source: 'git',
+                commit: commit.hash,
+                author: commit.author,
+                files: commit.files.slice(0, 10),
+            },
+        });
+    }
+    // Extract learnings from reverts (failed approaches)
+    for (const commit of reverts) {
+        memories.push({
+            id: `git-revert-${commit.hash.substring(0, 8)}`,
+            timestamp: commit.date,
+            type: 'thought',
+            content: `Failed approach reverted: ${commit.message}. This approach did not work - avoid similar changes.`,
+            tags: ['revert', 'failed-approach', ...extractFileTags(commit.files)],
+            importance: 8, // Reverts are always important learnings
+            metadata: {
+                source: 'git',
+                commit: commit.hash,
+                author: commit.author,
+                files: commit.files.slice(0, 10),
+            },
+        });
+    }
+    // Identify hot spots (frequently modified files)
+    const fileChangeCounts = new Map();
+    for (const commit of commits) {
+        for (const file of commit.files) {
+            fileChangeCounts.set(file, (fileChangeCounts.get(file) || 0) + 1);
+        }
+    }
+    const hotSpots = [...fileChangeCounts.entries()]
+        .sort((a, b) => b[1] - a[1])
+        .slice(0, 10)
+        .filter(([_, count]) => count >= 5);
+    if (hotSpots.length > 0) {
+        memories.push({
+            id: 'git-hotspots',
+            timestamp: new Date().toISOString(),
+            type: 'observation',
+            content: `Frequently modified files (hot spots): ${hotSpots.map(([f, c]) => `${f} (${c} changes)`).join(', ')}. These files may need extra attention during changes.`,
+            tags: ['hot-spots', 'code-quality'],
+            importance: 7,
+            metadata: {
+                source: 'git-analysis',
+                hotSpots: Object.fromEntries(hotSpots),
+            },
+        });
+    }
+    return memories;
+}
+// Helper functions
+function findMarkdownFiles(dir) {
+    const files = [];
+    const entries = readdirSync(dir, { withFileTypes: true });
+    for (const entry of entries) {
+        const fullPath = join(dir, entry.name);
+        if (entry.isDirectory() && !entry.name.startsWith('.')) {
+            files.push(...findMarkdownFiles(fullPath));
+        }
+        else if (entry.isFile() && (entry.name.endsWith('.md') || entry.name.endsWith('.mdx'))) {
+            files.push(fullPath);
+        }
+    }
+    return files;
+}
+function parseMarkdownFile(filePath, cwd, docType) {
+    try {
+        const content = readFileSync(filePath, 'utf-8');
+        const relativePath = relative(cwd, filePath);
+        // Extract title from first heading or filename
+        const titleMatch = content.match(/^#\s+(.+)$/m);
+        const title = titleMatch ? titleMatch[1] : basename(filePath, extname(filePath));
+        // Split content by headings
+        const sections = [];
+        const headingRegex = /^(#{1,3})\s+(.+)$/gm;
+        let lastIndex = 0;
+        let lastHeading = 'Introduction';
+        let match;
+        while ((match = headingRegex.exec(content)) !== null) {
+            // Save previous section
+            if (lastIndex > 0 || match.index > 0) {
+                const sectionContent = content.slice(lastIndex, match.index).trim();
+                if (sectionContent) {
+                    sections.push({
+                        heading: lastHeading,
+                        content: cleanMarkdown(sectionContent),
+                        type: categorizeSectionType(lastHeading, sectionContent, docType),
+                    });
+                }
+            }
+            lastHeading = match[2];
+            lastIndex = match.index + match[0].length;
+        }
+        // Add final section
+        const finalContent = content.slice(lastIndex).trim();
+        if (finalContent) {
+            sections.push({
+                heading: lastHeading,
+                content: cleanMarkdown(finalContent),
+                type: categorizeSectionType(lastHeading, finalContent, docType),
+            });
+        }
+        return {
+            path: relativePath,
+            title,
+            sections,
+        };
+    }
+    catch {
+        return null;
+    }
+}
+function cleanMarkdown(content) {
+    return content
+        .replace(/```[\s\S]*?```/g, '[code block]') // Replace code blocks
+        .replace(/`[^`]+`/g, (m) => m) // Keep inline code
+        .replace(/\[([^\]]+)\]\([^)]+\)/g, '$1') // Convert links to text
+        .replace(/!\[([^\]]*)\]\([^)]+\)/g, '[image: $1]') // Simplify images
+        .replace(/\n{3,}/g, '\n\n') // Normalize newlines
+        .trim();
+}
+function categorizeSectionType(heading, content, docType) {
+    const headingLower = heading.toLowerCase();
+    const contentLower = content.toLowerCase();
+    if (docType === 'adr')
+        return 'architecture';
+    if (headingLower.includes('install') ||
+        headingLower.includes('setup') ||
+        headingLower.includes('getting started') ||
+        headingLower.includes('quickstart') ||
+        headingLower.includes('prerequisites')) {
+        return 'setup';
+    }
+    if (headingLower.includes('architecture') ||
+        headingLower.includes('design') ||
+        headingLower.includes('structure') ||
+        headingLower.includes('overview')) {
+        return 'architecture';
+    }
+    if (headingLower.includes('api') ||
+        headingLower.includes('endpoint') ||
+        headingLower.includes('reference')) {
+        return 'api';
+    }
+    if (headingLower.includes('troubleshoot') ||
+        headingLower.includes('faq') ||
+        headingLower.includes('common issues') ||
+        headingLower.includes('known issues') ||
+        contentLower.includes('if you encounter')) {
+        return 'troubleshooting';
+    }
+    return 'general';
+}
+function mapSectionToMemoryType(sectionType) {
+    switch (sectionType) {
+        case 'architecture':
+            return 'thought';
+        case 'troubleshooting':
+            return 'observation';
+        default:
+            return 'observation';
+    }
+}
+function calculateDocImportance(section, filePath) {
+    let importance = 5;
+    // Boost README
+    if (filePath.toLowerCase().includes('readme'))
+        importance += 2;
+    // Boost architecture docs
+    if (section.type === 'architecture')
+        importance += 1;
+    // Boost setup/troubleshooting
+    if (section.type === 'setup' || section.type === 'troubleshooting')
+        importance += 1;
+    // ADRs are important
+    if (filePath.includes('adr') || filePath.includes('decision'))
+        importance += 2;
+    return Math.min(importance, 10);
+}
+function categorizeCommit(message) {
+    const msgLower = message.toLowerCase();
+    if (msgLower.startsWith('revert'))
+        return 'revert';
+    if (msgLower.startsWith('fix:') || msgLower.startsWith('fix(') || msgLower.includes('bugfix'))
+        return 'fix';
+    if (msgLower.startsWith('feat:') || msgLower.startsWith('feat(') || msgLower.startsWith('feature'))
+        return 'feat';
+    if (msgLower.startsWith('refactor:') || msgLower.startsWith('refactor('))
+        return 'refactor';
+    if (msgLower.startsWith('docs:') || msgLower.startsWith('docs('))
+        return 'docs';
+    if (msgLower.startsWith('test:') || msgLower.startsWith('test('))
+        return 'test';
+    if (msgLower.startsWith('chore:') || msgLower.startsWith('chore('))
+        return 'chore';
+    return 'other';
+}
+function extractBugFixLearning(commit) {
+    const message = commit.message;
+    const body = commit.body;
+    // Try to extract what was fixed
+    let learning = `Bug fixed: ${message}`;
+    if (body) {
+        // Look for "Fixes #123" or similar
+        const issueMatch = body.match(/(?:fixes|closes|resolves)\s+#(\d+)/i);
+        if (issueMatch) {
+            learning += ` (Issue #${issueMatch[1]})`;
+        }
+        // Add summary of body if present
+        const summary = summarizeBody(body);
+        if (summary) {
+            learning += `. ${summary}`;
+        }
+    }
+    // Add affected files context
+    if (commit.files.length > 0 && commit.files.length <= 5) {
+        learning += ` Affected files: ${commit.files.join(', ')}.`;
+    }
+    return learning;
+}
+function summarizeBody(body) {
+    // Take first meaningful line of body
+    const lines = body.split('\n').filter(l => l.trim() && !l.startsWith('Co-authored') && !l.startsWith('Signed-off'));
+    if (lines.length === 0)
+        return '';
+    const firstLine = lines[0].trim();
+    if (firstLine.length > 200) {
+        return firstLine.substring(0, 200) + '...';
+    }
+    return firstLine;
+}
+function calculateCommitImportance(commit) {
+    let importance = 5;
+    // Reverts are always important
+    if (commit.type === 'revert')
+        importance = 8;
+    // Large changes are more significant
+    if (commit.files.length > 10)
+        importance += 1;
+    // Changes to core files
+    const coreFiles = commit.files.filter(f => f.includes('config') ||
+        f.includes('package.json') ||
+        f.includes('index') ||
+        f.includes('main') ||
+        f.includes('app'));
+    if (coreFiles.length > 0)
+        importance += 1;
+    // Has detailed body
+    if (commit.body && commit.body.length > 50)
+        importance += 1;
+    return Math.min(importance, 10);
+}
+function extractFileTags(files) {
+    const tags = new Set();
+    for (const file of files.slice(0, 10)) {
+        // Extract directory as tag
+        const parts = file.split('/');
+        if (parts.length > 1) {
+            tags.add(parts[0]);
+        }
+        // Extract file type
+        const ext = extname(file).toLowerCase();
+        if (ext === '.ts' || ext === '.tsx')
+            tags.add('typescript');
+        if (ext === '.js' || ext === '.jsx')
+            tags.add('javascript');
+        if (ext === '.py')
+            tags.add('python');
+        if (ext === '.go')
+            tags.add('go');
+        if (ext === '.rs')
+            tags.add('rust');
+        if (ext === '.yml' || ext === '.yaml')
+            tags.add('config');
+        if (ext === '.json')
+            tags.add('config');
+        if (ext === '.md')
+            tags.add('docs');
+        // Special files
+        if (file.includes('test') || file.includes('spec'))
+            tags.add('tests');
+        if (file.includes('docker') || file.includes('Dockerfile'))
+            tags.add('docker');
+        if (file.includes('terraform') || file.endsWith('.tf'))
+            tags.add('terraform');
+    }
+    return [...tags].slice(0, 5);
+}
+function extractTags(section, filePath) {
+    const tags = [section.type];
+    // Add source file as tag
+    const fileName = basename(filePath, extname(filePath)).toLowerCase();
+    if (fileName !== 'readme') {
+        tags.push(fileName);
+    }
+    // Extract keywords from heading
+    const keywords = section.heading.toLowerCase()
+        .split(/[\s-_]+/)
+        .filter(w => w.length > 3 && !['the', 'and', 'for', 'with'].includes(w));
+    tags.push(...keywords.slice(0, 3));
+    return [...new Set(tags)].slice(0, 5);
+}
+function chunkContent(content, maxLength) {
+    if (content.length <= maxLength)
+        return [content];
+    const chunks = [];
+    const sentences = content.split(/(?<=[.!?])\s+/);
+    let currentChunk = '';
+    for (const sentence of sentences) {
+        if (currentChunk.length + sentence.length > maxLength) {
+            if (currentChunk)
+                chunks.push(currentChunk.trim());
+            currentChunk = sentence;
+        }
+        else {
+            currentChunk += (currentChunk ? ' ' : '') + sentence;
+        }
+    }
+    if (currentChunk)
+        chunks.push(currentChunk.trim());
+    return chunks;
+}
+/**
+ * Discover skills, droids, commands, and artifacts from various platforms
+ */
+export async function discoverSkillsAndArtifacts(cwd, verbose = false) {
+    const skills = [];
+    // Factory.ai droids, skills, and commands
+    const factoryPaths = [
+        { dir: '.factory/droids', type: 'droid' },
+        { dir: '.factory/skills', type: 'skill' },
+        { dir: '.factory/commands', type: 'command' },
+    ];
+    for (const { dir, type } of factoryPaths) {
+        const fullPath = join(cwd, dir);
+        if (existsSync(fullPath) && statSync(fullPath).isDirectory()) {
+            const files = readdirSync(fullPath);
+            for (const file of files) {
+                if (file.endsWith('.md') || file.endsWith('.yaml') || file.endsWith('.yml')) {
+                    const filePath = join(fullPath, file);
+                    const skill = parseSkillFile(filePath, cwd, type, 'factory');
+                    if (skill) {
+                        skills.push(skill);
+                        if (verbose)
+                            console.log(`  Found Factory ${type}: ${skill.name}`);
+                    }
+                }
+            }
+        }
+    }
+    // Claude Code agents and commands
+    const claudePaths = [
+        { dir: '.claude/agents', type: 'droid' },
+        { dir: '.claude/commands', type: 'command' },
+    ];
+    for (const { dir, type } of claudePaths) {
+        const fullPath = join(cwd, dir);
+        if (existsSync(fullPath) && statSync(fullPath).isDirectory()) {
+            const files = readdirSync(fullPath);
+            for (const file of files) {
+                if (file.endsWith('.md') || file.endsWith('.yaml') || file.endsWith('.yml')) {
+                    const filePath = join(fullPath, file);
+                    const skill = parseSkillFile(filePath, cwd, type, 'claude');
+                    if (skill) {
+                        skills.push(skill);
+                        if (verbose)
+                            console.log(`  Found Claude ${type}: ${skill.name}`);
+                    }
+                }
+            }
+        }
+    }
+    // OpenCode agents and commands
+    const opencodePaths = [
+        { dir: '.opencode/agent', type: 'droid' },
+        { dir: '.opencode/command', type: 'command' },
+    ];
+    for (const { dir, type } of opencodePaths) {
+        const fullPath = join(cwd, dir);
+        if (existsSync(fullPath) && statSync(fullPath).isDirectory()) {
+            const files = readdirSync(fullPath);
+            for (const file of files) {
+                if (file.endsWith('.md') || file.endsWith('.yaml') || file.endsWith('.yml') || file.endsWith('.json')) {
+                    const filePath = join(fullPath, file);
+                    const skill = parseSkillFile(filePath, cwd, type, 'opencode');
+                    if (skill) {
+                        skills.push(skill);
+                        if (verbose)
+                            console.log(`  Found OpenCode ${type}: ${skill.name}`);
+                    }
+                }
+            }
+        }
+    }
+    // Discover artifacts (reusable code patterns, templates)
+    const artifactDirs = [
+        'templates',
+        'snippets',
+        '.github/workflows',
+        'scripts',
+        'tools',
+    ];
+    for (const dir of artifactDirs) {
+        const fullPath = join(cwd, dir);
+        if (existsSync(fullPath) && statSync(fullPath).isDirectory()) {
+            const files = readdirSync(fullPath).slice(0, 20); // Limit to avoid too many
+            for (const file of files) {
+                const filePath = join(fullPath, file);
+                if (statSync(filePath).isFile()) {
+                    skills.push({
+                        name: basename(file, extname(file)),
+                        path: relative(cwd, filePath),
+                        type: 'artifact',
+                        description: `Reusable ${dir} artifact`,
+                        platform: 'generic',
+                    });
+                    if (verbose)
+                        console.log(`  Found artifact: ${file}`);
+                }
+            }
+        }
+    }
+    return skills;
+}
+/**
+ * Parse a skill/droid/command file to extract metadata
+ */
+function parseSkillFile(filePath, cwd, type, platform) {
+    try {
+        const content = readFileSync(filePath, 'utf-8');
+        const fileName = basename(filePath, extname(filePath));
+        let description = '';
+        // Try to extract description from different formats
+        if (filePath.endsWith('.md')) {
+            // Extract first paragraph or heading
+            const match = content.match(/^#\s+.+\n+(.+)/m);
+            if (match)
+                description = match[1].trim();
+        }
+        else if (filePath.endsWith('.yaml') || filePath.endsWith('.yml')) {
+            // Look for description field
+            const match = content.match(/description:\s*['"]?([^'"\n]+)/i);
+            if (match)
+                description = match[1].trim();
+        }
+        else if (filePath.endsWith('.json')) {
+            try {
+                const json = JSON.parse(content);
+                description = json.description || '';
+            }
+            catch {
+                // Ignore JSON parse errors
+            }
+        }
+        return {
+            name: fileName,
+            path: relative(cwd, filePath),
+            type,
+            description: description || `${platform} ${type}`,
+            platform,
+        };
+    }
+    catch {
+        return null;
+    }
+}
+/**
+ * Convert discovered skills into memory entries
+ */
+export function extractSkillMemories(skills) {
+    const memories = [];
+    const now = new Date().toISOString();
+    // Group by type
+    const droids = skills.filter(s => s.type === 'droid');
+    const skillItems = skills.filter(s => s.type === 'skill');
+    const commands = skills.filter(s => s.type === 'command');
+    const artifacts = skills.filter(s => s.type === 'artifact');
+    // Create memory for available droids/agents
+    if (droids.length > 0) {
+        memories.push({
+            id: 'skills-droids-available',
+            timestamp: now,
+            type: 'observation',
+            content: `Available AI agents/droids: ${droids.map(d => `${d.name} (${d.platform})`).join(', ')}. These can be invoked for specialized tasks.`,
+            tags: ['droids', 'agents', 'capabilities'],
+            importance: 8,
+            metadata: {
+                source: 'skills-discovery',
+                droids: droids.map(d => ({ name: d.name, platform: d.platform, path: d.path })),
+            },
+        });
+    }
+    // Create memory for available skills
+    if (skillItems.length > 0) {
+        memories.push({
+            id: 'skills-skills-available',
+            timestamp: now,
+            type: 'observation',
+            content: `Available skills: ${skillItems.map(s => s.name).join(', ')}. These provide specialized capabilities.`,
+            tags: ['skills', 'capabilities'],
+            importance: 7,
+            metadata: {
+                source: 'skills-discovery',
+                skills: skillItems.map(s => ({ name: s.name, platform: s.platform, path: s.path })),
+            },
+        });
+    }
+    // Create memory for available commands
+    if (commands.length > 0) {
+        memories.push({
+            id: 'skills-commands-available',
+            timestamp: now,
+            type: 'observation',
+            content: `Available slash commands: ${commands.map(c => `/${c.name}`).join(', ')}. Use these for quick actions.`,
+            tags: ['commands', 'shortcuts'],
+            importance: 7,
+            metadata: {
+                source: 'skills-discovery',
+                commands: commands.map(c => ({ name: c.name, platform: c.platform, path: c.path })),
+            },
+        });
+    }
+    // Create memories for individual droids with descriptions
+    for (const droid of droids) {
+        if (droid.description && droid.description.length > 20) {
+            memories.push({
+                id: `skill-droid-${droid.name}`,
+                timestamp: now,
+                type: 'thought',
+                content: `Droid "${droid.name}" (${droid.platform}): ${droid.description}. Located at ${droid.path}.`,
+                tags: ['droid', droid.platform, droid.name],
+                importance: 6,
+                metadata: {
+                    source: 'skills-discovery',
+                    ...droid,
+                },
+            });
+        }
+    }
+    // Create memory for useful artifacts/templates
+    const usefulArtifacts = artifacts.filter(a => a.path.includes('workflow') ||
+        a.path.includes('template') ||
+        a.path.includes('script'));
+    if (usefulArtifacts.length > 0) {
+        memories.push({
+            id: 'skills-artifacts-available',
+            timestamp: now,
+            type: 'observation',
+            content: `Useful project artifacts: ${usefulArtifacts.slice(0, 10).map(a => a.path).join(', ')}. These can be referenced or reused.`,
+            tags: ['artifacts', 'templates', 'reusable'],
+            importance: 5,
+            metadata: {
+                source: 'skills-discovery',
+                artifacts: usefulArtifacts.map(a => a.path),
+            },
+        });
+    }
+    return memories;
+}
+//# sourceMappingURL=prepopulate.js.map