@kbdavis07/openspec 1.5.0-nov.1

package/dist/core/project-config.js

@@ -0,0 +1,373 @@
+import { existsSync, readFileSync, statSync } from 'fs';
+import path from 'path';
+import { parse as parseYaml } from 'yaml';
+import { z } from 'zod';
+/**
+ * Zod schema for project configuration.
+ *
+ * Purpose:
+ * 1. Documentation - clearly defines the config file structure
+ * 2. Type safety - TypeScript infers ProjectConfig type from schema
+ * 3. Runtime validation - uses safeParse() for resilient field-by-field validation
+ *
+ * Why Zod over manual validation:
+ * - Helps understand OpenSpec's data interfaces at a glance
+ * - Single source of truth for type and validation
+ * - Consistent with other OpenSpec schemas
+ */
+export const ProjectConfigSchema = z.object({
+    // Required: which schema to use (e.g., "spec-driven", or project-local schema name)
+    schema: z
+        .string()
+        .min(1)
+        .describe('The workflow schema to use (e.g., "spec-driven")'),
+    // Optional: project context (injected into all artifact instructions)
+    // Max size: 50KB (enforced during parsing)
+    context: z
+        .string()
+        .optional()
+        .describe('Project context injected into all artifact instructions'),
+    // Optional: per-artifact rules (additive to schema's built-in guidance)
+    rules: z
+        .record(z.string(), // artifact ID
+    z.array(z.string()) // list of rules
+    )
+        .optional()
+        .describe('Per-artifact rules, keyed by artifact ID'),
+    // Note: the `references` field (id strings or {id, remote} maps) is
+    // deliberately absent here — readProjectConfig parses and normalizes
+    // it by hand (see DeclarationEntry below); a schema entry nothing
+    // parses would only drift from the real behavior.
+    // Optional: the declared default store. Only consulted by root
+    // resolution when this openspec/ directory is config-only (no specs/
+    // or changes/); a fallback, never an override.
+    store: z
+        .string()
+        .optional()
+        .describe('Store id used as the OpenSpec root when no local planning shape exists'),
+});
+/**
+ * Parser for `references:` declarations: string entries or
+ * {id, remote} maps, normalized to DeclarationEntry[]. Dedup keys on
+ * id and keeps the first position; the first entry carrying a remote
+ * supplies it (a later duplicate fills a missing remote, never
+ * overrides). Invalid entries drop with a warning like other resilient
+ * fields; returns undefined when the field is absent or normalizes to
+ * empty.
+ */
+function parseDeclarationList(raw) {
+    const fieldName = 'references';
+    if (raw === undefined) {
+        return undefined;
+    }
+    if (!Array.isArray(raw)) {
+        console.warn(`Invalid '${fieldName}' field in config (must be an array of store ids)`);
+        return undefined;
+    }
+    const byId = new Map();
+    let droppedEntries = false;
+    let droppedRemotes = false;
+    for (const entry of raw) {
+        let declaration = null;
+        if (typeof entry === 'string') {
+            declaration = { id: entry };
+        }
+        else if (entry && typeof entry === 'object' && !Array.isArray(entry)) {
+            const candidate = entry;
+            if (typeof candidate.id === 'string') {
+                declaration = { id: candidate.id };
+                if (typeof candidate.remote === 'string' && candidate.remote.length > 0) {
+                    declaration.remote = candidate.remote;
+                }
+                else if (candidate.remote !== undefined) {
+                    droppedRemotes = true; // remote dropped, id kept
+                }
+            }
+        }
+        if (!declaration) {
+            droppedEntries = true;
+            continue;
+        }
+        const existing = byId.get(declaration.id);
+        if (!existing) {
+            byId.set(declaration.id, declaration);
+        }
+        else if (existing.remote === undefined && declaration.remote !== undefined) {
+            existing.remote = declaration.remote;
+        }
+    }
+    if (droppedEntries) {
+        console.warn(`Some '${fieldName}' entries are invalid, ignoring them`);
+    }
+    if (droppedRemotes) {
+        console.warn(`Some '${fieldName}' remotes are not non-empty strings; the ids are kept without a clone source`);
+    }
+    return byId.size > 0 ? [...byId.values()] : undefined;
+}
+export const MAX_CONTEXT_SIZE = 50 * 1024; // 50KB hard limit, shared with the references index
+/**
+ * Read and parse openspec/config.yaml from project root.
+ * Uses resilient parsing - validates each field independently using Zod safeParse.
+ * Returns null if file doesn't exist.
+ * Returns partial config if some fields are invalid (with warnings).
+ *
+ * Performance note (Jan 2025):
+ * Benchmarks showed direct file reads are fast enough without caching:
+ * - Typical config (1KB): ~0.5ms per read
+ * - Large config (50KB): ~1.6ms per read
+ * - Missing config: ~0.01ms per read
+ * Config is read 1-2 times per command (schema resolution + instruction loading),
+ * adding ~1-3ms total overhead. Caching would add complexity (mtime checks,
+ * invalidation logic) for negligible benefit. Direct reads also ensure config
+ * changes are reflected immediately without stale cache issues.
+ *
+ * @param projectRoot - The root directory of the project (where `openspec/` lives)
+ * @returns Parsed config or null if file doesn't exist
+ */
+export function readProjectConfig(projectRoot) {
+    const configPath = resolveConfigFilePath(projectRoot);
+    if (configPath === null) {
+        return null; // No config is OK
+    }
+    try {
+        const content = readFileSync(configPath, 'utf-8');
+        const raw = parseYaml(content);
+        if (!raw || typeof raw !== 'object') {
+            console.warn(`openspec/config.yaml is not a valid YAML object`);
+            return null;
+        }
+        const config = {};
+        // Parse schema field using Zod
+        const schemaField = z.string().min(1);
+        const schemaResult = schemaField.safeParse(raw.schema);
+        if (schemaResult.success) {
+            config.schema = schemaResult.data;
+        }
+        else if (raw.schema !== undefined) {
+            console.warn(`Invalid 'schema' field in config (must be non-empty string)`);
+        }
+        // Parse context field with size limit
+        if (raw.context !== undefined) {
+            const contextField = z.string();
+            const contextResult = contextField.safeParse(raw.context);
+            if (contextResult.success) {
+                const contextSize = Buffer.byteLength(contextResult.data, 'utf-8');
+                if (contextSize > MAX_CONTEXT_SIZE) {
+                    console.warn(`Context too large (${(contextSize / 1024).toFixed(1)}KB, limit: ${MAX_CONTEXT_SIZE / 1024}KB)`);
+                    console.warn(`Ignoring context field`);
+                }
+                else {
+                    config.context = contextResult.data;
+                }
+            }
+            else {
+                console.warn(`Invalid 'context' field in config (must be string)`);
+            }
+        }
+        // Parse rules field using Zod
+        if (raw.rules !== undefined) {
+            const rulesField = z.record(z.string(), z.array(z.string()));
+            // First check if it's an object structure (guard against null since typeof null === 'object')
+            if (typeof raw.rules === 'object' && raw.rules !== null && !Array.isArray(raw.rules)) {
+                const parsedRules = {};
+                let hasValidRules = false;
+                for (const [artifactId, rules] of Object.entries(raw.rules)) {
+                    const rulesArrayResult = z.array(z.string()).safeParse(rules);
+                    if (rulesArrayResult.success) {
+                        // Filter out empty strings
+                        const validRules = rulesArrayResult.data.filter((r) => r.length > 0);
+                        if (validRules.length > 0) {
+                            parsedRules[artifactId] = validRules;
+                            hasValidRules = true;
+                        }
+                        if (validRules.length < rulesArrayResult.data.length) {
+                            console.warn(`Some rules for '${artifactId}' are empty strings, ignoring them`);
+                        }
+                    }
+                    else {
+                        console.warn(`Rules for '${artifactId}' must be an array of strings, ignoring this artifact's rules`);
+                    }
+                }
+                if (hasValidRules) {
+                    config.rules = parsedRules;
+                }
+            }
+            else {
+                console.warn(`Invalid 'rules' field in config (must be object)`);
+            }
+        }
+        const references = parseDeclarationList(raw.references);
+        if (references) {
+            config.references = references;
+        }
+        // Parse store pointer field: a string, or dropped with a warning.
+        // (Root resolution does NOT use this parse — it uses readStorePointer
+        // below, which errors on malformed pointers instead of dropping.)
+        if (raw.store !== undefined) {
+            if (typeof raw.store === 'string') {
+                config.store = raw.store;
+            }
+            else {
+                console.warn(`Warning: ignoring invalid store: field in ${configPathForWarnings(projectRoot)} (must be a single store id string).`);
+            }
+        }
+        // Return partial config even if some fields failed
+        return Object.keys(config).length > 0 ? config : null;
+    }
+    catch (error) {
+        console.warn(`Warning: could not parse ${configPathForWarnings(projectRoot)} (${error instanceof Error ? error.message.split('\n')[0] : String(error)}); ignoring it.`);
+        return null;
+    }
+}
+function configPathForWarnings(projectRoot) {
+    return resolveConfigFilePath(projectRoot) ?? path.join(projectRoot, 'openspec', 'config.yaml');
+}
+/**
+ * Validate artifact IDs in rules against a schema's artifacts.
+ * Called during instruction loading (when schema is known).
+ * Returns warnings for unknown artifact IDs.
+ *
+ * @param rules - The rules object from config
+ * @param validArtifactIds - Set of valid artifact IDs from the schema
+ * @param schemaName - Name of the schema for error messages
+ * @returns Array of warning messages for unknown artifact IDs
+ */
+export function validateConfigRules(rules, validArtifactIds, schemaName) {
+    const warnings = [];
+    for (const artifactId of Object.keys(rules)) {
+        if (!validArtifactIds.has(artifactId)) {
+            const validIds = Array.from(validArtifactIds).sort().join(', ');
+            warnings.push(`Unknown artifact ID in rules: "${artifactId}". ` +
+                `Valid IDs for schema "${schemaName}": ${validIds}`);
+        }
+    }
+    return warnings;
+}
+/**
+ * Suggest valid schema names when user provides invalid schema.
+ * Uses fuzzy matching to find similar names.
+ *
+ * @param invalidSchemaName - The invalid schema name from config
+ * @param availableSchemas - List of available schemas with their type (built-in or project-local)
+ * @returns Error message with suggestions and available schemas
+ */
+export function suggestSchemas(invalidSchemaName, availableSchemas) {
+    // Simple fuzzy match: Levenshtein distance
+    function levenshtein(a, b) {
+        const matrix = [];
+        for (let i = 0; i <= b.length; i++) {
+            matrix[i] = [i];
+        }
+        for (let j = 0; j <= a.length; j++) {
+            matrix[0][j] = j;
+        }
+        for (let i = 1; i <= b.length; i++) {
+            for (let j = 1; j <= a.length; j++) {
+                if (b.charAt(i - 1) === a.charAt(j - 1)) {
+                    matrix[i][j] = matrix[i - 1][j - 1];
+                }
+                else {
+                    matrix[i][j] = Math.min(matrix[i - 1][j - 1] + 1, matrix[i][j - 1] + 1, matrix[i - 1][j] + 1);
+                }
+            }
+        }
+        return matrix[b.length][a.length];
+    }
+    // Find closest matches (distance <= 3)
+    const suggestions = availableSchemas
+        .map((s) => ({ ...s, distance: levenshtein(invalidSchemaName, s.name) }))
+        .filter((s) => s.distance <= 3)
+        .sort((a, b) => a.distance - b.distance)
+        .slice(0, 3);
+    const builtIn = availableSchemas.filter((s) => s.isBuiltIn).map((s) => s.name);
+    const projectLocal = availableSchemas.filter((s) => !s.isBuiltIn).map((s) => s.name);
+    let message = `Schema '${invalidSchemaName}' not found in openspec/config.yaml\n\n`;
+    if (suggestions.length > 0) {
+        message += `Did you mean one of these?\n`;
+        suggestions.forEach((s) => {
+            const type = s.isBuiltIn ? 'built-in' : 'project-local';
+            message += `  - ${s.name} (${type})\n`;
+        });
+        message += '\n';
+    }
+    message += `Available schemas:\n`;
+    if (builtIn.length > 0) {
+        message += `  Built-in: ${builtIn.join(', ')}\n`;
+    }
+    if (projectLocal.length > 0) {
+        message += `  Project-local: ${projectLocal.join(', ')}\n`;
+    }
+    else {
+        message += `  Project-local: (none found)\n`;
+    }
+    message += `\nFix: Edit openspec/config.yaml and change 'schema: ${invalidSchemaName}' to a valid schema name`;
+    return message;
+}
+/**
+ * Warning-silent targeted read of the `store:` pointer. Used by root
+ * resolution (which must not re-emit the resilient parser's field
+ * warnings) and by `openspec init`'s pointer guard. Unlike
+ * `readProjectConfig`, a malformed value is REPORTED, not dropped —
+ * a dropped pointer would silently flip where work lands.
+ */
+export function readStorePointer(projectRoot) {
+    const configPath = resolveConfigFilePath(projectRoot);
+    if (configPath === null) {
+        return { filePath: null };
+    }
+    try {
+        const raw = parseYaml(readFileSync(configPath, 'utf-8'));
+        // Empty, comments-only, or non-mapping configs carry no pointer;
+        // they are imperfect, not malformed (readProjectConfig owns the
+        // field warnings for those).
+        if (!raw || typeof raw !== 'object' || Array.isArray(raw)) {
+            return { filePath: configPath };
+        }
+        const value = raw.store;
+        if (value === undefined) {
+            return { filePath: configPath };
+        }
+        if (typeof value === 'string') {
+            return { value, filePath: configPath };
+        }
+        return { malformed: 'non_string', filePath: configPath };
+    }
+    catch {
+        return { malformed: 'unparseable', filePath: configPath };
+    }
+}
+/** Shared .yaml/.yml probe used by readProjectConfig and readStorePointer. */
+export function resolveConfigFilePath(projectRoot) {
+    const yamlPath = path.join(projectRoot, 'openspec', 'config.yaml');
+    if (existsSync(yamlPath)) {
+        return yamlPath;
+    }
+    const ymlPath = path.join(projectRoot, 'openspec', 'config.yml');
+    return existsSync(ymlPath) ? ymlPath : null;
+}
+/** Human rendering of a malformed pointer reason, shared by every surface. */
+export function storePointerProblem(reason) {
+    return reason === 'unparseable'
+        ? 'the config file could not be read as YAML'
+        : 'the store key must be a single store id string';
+}
+/**
+ * One classification for "real root vs config-only pointer dir", shared
+ * by root resolution and the init pointer guard so they can never
+ * disagree (slice 3.2).
+ */
+export function classifyOpenSpecDir(projectRoot) {
+    const openspecDir = path.join(projectRoot, 'openspec');
+    const hasPlanningShape = isDirectorySync(path.join(openspecDir, 'specs')) ||
+        isDirectorySync(path.join(openspecDir, 'changes'));
+    return { hasPlanningShape, pointer: readStorePointer(projectRoot) };
+}
+function isDirectorySync(candidatePath) {
+    try {
+        return statSync(candidatePath).isDirectory();
+    }
+    catch {
+        return false;
+    }
+}
+//# sourceMappingURL=project-config.js.map

package/dist/core/references.d.ts

@@ -0,0 +1,63 @@
+import { type StoreDiagnostic } from './store/errors.js';
+import { listStoreRegistryEntries } from './store/foundation.js';
+import { type ResolvedOpenSpecRoot } from './root-selection.js';
+import { type DeclarationEntry } from './project-config.js';
+export interface ReferenceSpecEntry {
+    id: string;
+    summary: string;
+}
+export interface ReferenceIndexEntry {
+    store_id: string;
+    root?: string;
+    specs?: ReferenceSpecEntry[];
+    fetch?: string;
+    status: StoreDiagnostic[];
+}
+/**
+ * Tolerant first-Purpose-line extraction. parseSpec() throws on specs
+ * without Purpose/Requirements sections; the index must never fail on an
+ * imperfect upstream spec, so this scans for the heading directly —
+ * fence-aware, so `## Purpose` inside a code block never matches, and
+ * tolerant of CommonMark closing hashes (`## Purpose ##`).
+ */
+export declare function extractFirstPurposeLine(markdown: string): string;
+export declare function fetchRecipe(storeId: string): string;
+/**
+ * Pure renderer for the artifact-instructions XML block. Also the byte
+ * budget's measuring stick (it is the larger rendering).
+ */
+export declare function renderReferencedStoresBlock(entries: ReferenceIndexEntry[]): string;
+/** Pure renderer for the apply-instructions markdown section. */
+export declare function renderReferencedStoresSection(entries: ReferenceIndexEntry[]): string;
+/**
+ * Strings rendered into agent guidance can come from cloned content
+ * (spec directory names, Purpose lines, config-declared remotes). One
+ * line in, one line out: control characters and newlines must never
+ * let hostile content forge instruction lines (slice 6.1 hardening).
+ */
+export declare function sanitizeInline(value: string, maxLength?: number): string;
+export interface AssembleReferenceIndexInput {
+    references: DeclarationEntry[];
+    resolvedRoot: ResolvedOpenSpecRoot;
+    globalDataDir?: string;
+    /**
+     * Health mode (3.6): false skips the spec-file reads AND the byte
+     * budget — entries carry no `specs`/`fetch` keys at all, and the
+     * content-only truncation diagnostic can never appear.
+     */
+    includeSpecs?: boolean;
+    /**
+     * Pre-read registry entries (3.6): `[]` = registry empty or absent,
+     * `null` = unreadable, undefined = read internally as before.
+     * (Mirrors the internal post-read variable — never inject a raw
+     * read result: a healthy-absent registry reads as null.)
+     */
+    registryEntries?: ReturnType<typeof listStoreRegistryEntries> | null;
+}
+/**
+ * Builds the referenced-store index. One registry read per call; one
+ * level deep (a referenced store's own references are never followed);
+ * self-references omitted; every failure degrades to a warning entry.
+ */
+export declare function assembleReferenceIndex(input: AssembleReferenceIndexInput): Promise<ReferenceIndexEntry[]>;
+//# sourceMappingURL=references.d.ts.map