birdclaw 0.8.2 → 0.8.3

package/CHANGELOG.md

@@ -1,5 +1,21 @@
 # CHANGELOG
 
+## 0.8.3 - 2026-06-15
+
+### Changed
+
+- Centralize SQLite writes and migrations, unify live-sync ingestion, and move retained web server state to TanStack Query for faster sidebar navigation and fewer duplicate reads.
+- Isolate web reads onto query-only SQLite connections while serializing action and sync writes through one measured writer queue.
+- Share page caps, cursor progression, delays, and repeated-cursor protection across live timeline, DM, and follow-graph sync.
+- Stream archive arrays and backup JSONL records through resumable batches instead of buffering entire source files.
+- Split query access into typed timeline, DM, status, resource, and action modules while retaining the existing compatibility import.
+- Retain Blocks, Inbox, Data Sources, and Rate Limits server state in TanStack Query across sidebar navigation and targeted refreshes.
+- Split compose, inbox, follow-graph, database, and backup CLI registration into domain-owned command modules.
+
+### Fixed
+
+- Preload hover-preview avatars after page load so citation and profile cards open with cached thumbnails.
+
 ## 0.8.2 - 2026-06-15
 
 ### Fixed

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "birdclaw",
-	"version": "0.8.2",
+	"version": "0.8.3",
 	"description": "Local Twitter memory in SQLite for archives, DMs, likes, bookmarks, and moderation",
 	"homepage": "https://github.com/steipete/birdclaw#readme",
 	"license": "MIT",
@@ -54,6 +54,7 @@
 	"dependencies": {
 		"@steipete/sweet-cookie": "^0.4.0",
 		"@tailwindcss/vite": "^4.3.0",
+		"@tanstack/react-query": "^5.101.0",
 		"@tanstack/react-router": "1.170.15",
 		"@tanstack/react-router-ssr-query": "1.167.1",
 		"@tanstack/react-start": "1.168.25",

package/src/cli/command-context.ts

@@ -0,0 +1,17 @@
+import type { Command } from "commander";
+
+export interface CliCommandContext {
+	program: Command;
+	print: (data: unknown, asJson: boolean) => void;
+	asJson: () => boolean;
+	autoSyncAfterWrite: () => Promise<void>;
+	autoUpdateBeforeRead: () => Promise<void>;
+	parseNonNegativeIntegerOption: (
+		value: string | undefined,
+		option: string,
+	) => number | undefined;
+	parsePositiveIntegerOption: (
+		value: string | undefined,
+		option: string,
+	) => number | undefined;
+}

package/src/cli/register-analysis.ts

@@ -0,0 +1,500 @@
+import type { CliCommandContext } from "./command-context";
+import {
+	normalizeDigestLanguage,
+	streamPeriodDigest,
+	type PeriodDigestOptions,
+	type PeriodDigestPreset,
+} from "#/lib/period-digest";
+import {
+	streamProfileAnalysis,
+	type ProfileAnalysisOptions,
+} from "#/lib/profile-analysis";
+import { runResearchMode } from "#/lib/research";
+import {
+	streamSearchDiscussion,
+	type SearchDiscussionOptions,
+	type SearchDiscussionSource,
+} from "#/lib/search-discussion";
+
+export function registerAnalysisCommands({
+	program,
+	print,
+	autoUpdateBeforeRead,
+	parseNonNegativeIntegerOption,
+	parsePositiveIntegerOption,
+}: CliCommandContext) {
+	function printError(error: string) {
+		console.error(JSON.stringify({ error }));
+	}
+
+	function parseDigestLiveModeOption(
+		value: string | undefined,
+	): PeriodDigestOptions["liveSyncMode"] {
+		const normalized = (value ?? "xurl").trim().toLowerCase();
+		if (
+			normalized === "auto" ||
+			normalized === "bird" ||
+			normalized === "xurl"
+		) {
+			return normalized;
+		}
+		printError("--live-mode must be auto, bird, or xurl");
+		process.exitCode = 1;
+		return undefined;
+	}
+	function parseDigestPeriod(value: string | undefined): PeriodDigestPreset {
+		const normalized = value?.trim().toLowerCase();
+		if (normalized === "yesterday") return "yesterday";
+		if (normalized === "24h" || normalized === "day") return "24h";
+		if (normalized === "week" || normalized === "7d") return "week";
+		return "today";
+	}
+
+	function buildDigestOptions(
+		period: string | undefined,
+		options: {
+			account?: string;
+			includeDms?: boolean;
+			model?: string;
+			language?: string;
+			refresh?: boolean;
+			since?: string;
+			until?: string;
+			maxTweets?: string;
+			maxLinks?: string;
+			liveSync?: boolean;
+			liveMode?: string;
+		},
+	): PeriodDigestOptions | null {
+		const maxTweets = parseNonNegativeIntegerOption(
+			options.maxTweets,
+			"--max-tweets",
+		);
+		if (options.maxTweets !== undefined && maxTweets === undefined) {
+			return null;
+		}
+		const maxLinks = parseNonNegativeIntegerOption(
+			options.maxLinks,
+			"--max-links",
+		);
+		if (options.maxLinks !== undefined && maxLinks === undefined) {
+			return null;
+		}
+		const liveSyncMode = parseDigestLiveModeOption(options.liveMode);
+		if (liveSyncMode === undefined) {
+			return null;
+		}
+		let language: string | undefined;
+		try {
+			language = normalizeDigestLanguage(options.language);
+		} catch (error) {
+			printError(error instanceof Error ? error.message : String(error));
+			process.exitCode = 1;
+			return null;
+		}
+		return {
+			period: parseDigestPeriod(period),
+			since: options.since,
+			until: options.until,
+			account: options.account,
+			includeDms: Boolean(options.includeDms),
+			refresh: Boolean(options.refresh),
+			model: options.model,
+			language,
+			maxTweets,
+			maxLinks,
+			liveSync: options.liveSync !== false,
+			liveSyncMode,
+		};
+	}
+
+	function runDigestCli(options: PeriodDigestOptions) {
+		const asJson = Boolean(program.opts().json);
+		return streamPeriodDigest(options, {
+			onDelta: asJson
+				? undefined
+				: (delta) => {
+						process.stdout.write(delta);
+					},
+		}).then((result) => {
+			if (asJson) {
+				print(result, true);
+				return;
+			}
+			if (!result.markdown.endsWith("\n")) {
+				process.stdout.write("\n");
+			}
+		});
+	}
+
+	function parseSearchDiscussionSource(
+		value: string | undefined,
+	): SearchDiscussionSource | undefined {
+		const normalized = (value ?? "all").trim().toLowerCase();
+		if (
+			normalized === "all" ||
+			normalized === "home" ||
+			normalized === "mentions" ||
+			normalized === "authored" ||
+			normalized === "search" ||
+			normalized === "likes" ||
+			normalized === "bookmarks"
+		) {
+			return normalized;
+		}
+		printError(
+			"--source must be all, search, home, mentions, authored, likes, or bookmarks",
+		);
+		process.exitCode = 1;
+		return undefined;
+	}
+
+	function parseTweetSearchMode(value: string | undefined) {
+		const normalized = (value ?? "auto").trim().toLowerCase();
+		if (
+			normalized === "auto" ||
+			normalized === "bird" ||
+			normalized === "xurl" ||
+			normalized === "local"
+		) {
+			return normalized;
+		}
+		printError("--mode must be auto, bird, xurl, or local");
+		process.exitCode = 1;
+		return undefined;
+	}
+
+	function buildSearchDiscussionOptions(
+		query: string,
+		options: {
+			account?: string;
+			source?: string;
+			includeDms?: boolean;
+			since?: string;
+			until?: string;
+			question?: string;
+			originalsOnly?: boolean;
+			hideLowQuality?: boolean;
+			mode?: string;
+			model?: string;
+			refresh?: boolean;
+			limit?: string;
+			maxPages?: string;
+		},
+	): SearchDiscussionOptions | null {
+		const source = parseSearchDiscussionSource(options.source);
+		if (!source) return null;
+		const mode = parseTweetSearchMode(options.mode);
+		if (!mode) return null;
+		const limit = parsePositiveIntegerOption(options.limit, "--limit");
+		if (options.limit !== undefined && limit === undefined) {
+			return null;
+		}
+		const maxPages = parsePositiveIntegerOption(
+			options.maxPages,
+			"--max-pages",
+		);
+		if (options.maxPages !== undefined && maxPages === undefined) {
+			return null;
+		}
+		return {
+			query,
+			account: options.account,
+			source,
+			includeDms: Boolean(options.includeDms),
+			since: options.since,
+			until: options.until,
+			question: options.question,
+			originalsOnly: Boolean(options.originalsOnly),
+			hideLowQuality: Boolean(options.hideLowQuality),
+			mode,
+			model: options.model,
+			refresh: Boolean(options.refresh),
+			limit,
+			maxPages,
+		};
+	}
+
+	function runSearchDiscussionCli(options: SearchDiscussionOptions) {
+		const asJson = Boolean(program.opts().json);
+		return streamSearchDiscussion(options, {
+			onDelta: asJson
+				? undefined
+				: (delta) => {
+						process.stdout.write(delta);
+					},
+		}).then((result) => {
+			if (asJson) {
+				print(result, true);
+				return;
+			}
+			if (!result.markdown.endsWith("\n")) {
+				process.stdout.write("\n");
+			}
+		});
+	}
+
+	function buildProfileAnalysisOptions(
+		handle: string,
+		options: {
+			account?: string;
+			model?: string;
+			refresh?: boolean;
+			maxTweets?: string;
+			maxPages?: string;
+			maxConversations?: string;
+			maxConversationPages?: string;
+			conversationDelayMs?: string;
+			rateLimitRetryMs?: string;
+			rateLimitRetries?: string;
+		},
+	): ProfileAnalysisOptions | null {
+		const maxTweets = parsePositiveIntegerOption(
+			options.maxTweets,
+			"--max-tweets",
+		);
+		if (options.maxTweets !== undefined && maxTweets === undefined) {
+			return null;
+		}
+		const maxPages = parsePositiveIntegerOption(
+			options.maxPages,
+			"--max-pages",
+		);
+		if (options.maxPages !== undefined && maxPages === undefined) {
+			return null;
+		}
+		const maxConversations = parsePositiveIntegerOption(
+			options.maxConversations,
+			"--max-conversations",
+		);
+		if (
+			options.maxConversations !== undefined &&
+			maxConversations === undefined
+		) {
+			return null;
+		}
+		const maxConversationPages = parsePositiveIntegerOption(
+			options.maxConversationPages,
+			"--max-conversation-pages",
+		);
+		if (
+			options.maxConversationPages !== undefined &&
+			maxConversationPages === undefined
+		) {
+			return null;
+		}
+		const conversationDelayMs = parseNonNegativeIntegerOption(
+			options.conversationDelayMs,
+			"--conversation-delay-ms",
+		);
+		if (
+			options.conversationDelayMs !== undefined &&
+			conversationDelayMs === undefined
+		) {
+			return null;
+		}
+		const rateLimitRetryMs = parseNonNegativeIntegerOption(
+			options.rateLimitRetryMs,
+			"--rate-limit-retry-ms",
+		);
+		if (
+			options.rateLimitRetryMs !== undefined &&
+			rateLimitRetryMs === undefined
+		) {
+			return null;
+		}
+		const rateLimitMaxRetries = parseNonNegativeIntegerOption(
+			options.rateLimitRetries,
+			"--rate-limit-retries",
+		);
+		if (
+			options.rateLimitRetries !== undefined &&
+			rateLimitMaxRetries === undefined
+		) {
+			return null;
+		}
+		return {
+			handle,
+			account: options.account,
+			model: options.model,
+			refresh: Boolean(options.refresh),
+			maxTweets,
+			maxPages,
+			maxConversations,
+			maxConversationPages,
+			conversationDelayMs,
+			rateLimitRetryMs,
+			rateLimitMaxRetries,
+		};
+	}
+
+	function runProfileAnalysisCli(options: ProfileAnalysisOptions) {
+		const asJson = Boolean(program.opts().json);
+		return streamProfileAnalysis(options, {
+			onDelta: asJson
+				? undefined
+				: (delta) => {
+						process.stdout.write(delta);
+					},
+			onEvent: asJson
+				? undefined
+				: (event) => {
+						if (event.type === "status") {
+							process.stderr.write(
+								event.detail
+									? `${event.label}: ${event.detail}\n`
+									: `${event.label}\n`,
+							);
+						}
+					},
+		}).then((result) => {
+			if (asJson) {
+				print(result, true);
+				return;
+			}
+			if (!result.markdown.endsWith("\n")) {
+				process.stdout.write("\n");
+			}
+		});
+	}
+	program
+		.command("research [query]")
+		.description("Build a markdown research brief from bookmarked threads")
+		.option("--account <accountId>", "Account id")
+		.option("--limit <n>", "Seed bookmark limit", "20")
+		.option("--thread-depth <n>", "Maximum ancestor walk depth", "10")
+		.option("--out <path>", "Write the markdown brief to a file")
+		.action(async (query, options) => {
+			await autoUpdateBeforeRead();
+			const report = await runResearchMode({
+				account: options.account,
+				query,
+				limit: Number(options.limit),
+				maxThreadDepth: Number(options.threadDepth),
+				outPath: options.out,
+			});
+			print(
+				program.opts().json ? report : report.markdown,
+				program.opts().json ?? false,
+			);
+		});
+
+	program
+		.command("discuss <query>")
+		.description("Search live/local tweets and summarize the results with AI")
+		.option("--account <accountId>", "Account id")
+		.option(
+			"--source <source>",
+			"all, search, home, mentions, authored, likes, or bookmarks",
+			"search",
+		)
+		.option("--mode <mode>", "auto, bird, xurl, or local", "xurl")
+		.option("--include-dms", "Include private DM search matches")
+		.option(
+			"--since <isoDate>",
+			"Include matches created at or after this date",
+		)
+		.option("--until <isoDate>", "Include matches created before this date")
+		.option("--question <prompt>", "Discussion question or angle")
+		.option("--originals-only", "Exclude authored replies that start with @")
+		.option("--hide-low-quality", "Hide RTs, tiny replies, and link-only noise")
+		.option("--model <model>", "OpenAI model id")
+		.option("--refresh", "Bypass the local discussion cache")
+		.option("--limit <n>", "Maximum tweet context", "20000")
+		.option("--max-pages <n>", "Maximum live search pages", "200")
+		.action(async (query, options) => {
+			await autoUpdateBeforeRead();
+			const discussionOptions = buildSearchDiscussionOptions(query, options);
+			if (!discussionOptions) return;
+			await runSearchDiscussionCli(discussionOptions);
+		});
+
+	program
+		.command("profile-analyze <handle>")
+		.alias("profile-analyse")
+		.description("Backfill a profile with xurl and summarize it with AI")
+		.option("--account <accountId>", "Account id")
+		.option("--model <model>", "OpenAI model id")
+		.option("--refresh", "Bypass profile fetch and analysis caches")
+		.option("--max-tweets <n>", "Maximum profile tweets", "10000")
+		.option("--max-pages <n>", "Maximum profile timeline pages", "100")
+		.option(
+			"--max-conversations <n>",
+			"Maximum conversations to backfill",
+			"80",
+		)
+		.option(
+			"--max-conversation-pages <n>",
+			"Maximum pages per conversation",
+			"3",
+		)
+		.option(
+			"--conversation-delay-ms <n>",
+			"Delay between conversation search calls",
+		)
+		.option(
+			"--rate-limit-retry-ms <n>",
+			"Delay before retrying conversation 429s",
+		)
+		.option("--rate-limit-retries <n>", "Conversation 429 retry count")
+		.action(async (handle, options) => {
+			await autoUpdateBeforeRead();
+			const analysisOptions = buildProfileAnalysisOptions(handle, options);
+			if (!analysisOptions) return;
+			await runProfileAnalysisCli(analysisOptions);
+		});
+
+	program
+		.command("today")
+		.description("Stream an AI digest of what happened today")
+		.option("--account <accountId>", "Account id")
+		.option("--include-dms", "Include private DM context")
+		.option("--model <model>", "OpenAI model id")
+		.option(
+			"--language <tag>",
+			"Report language as a Unicode locale id, e.g. zh-CN (env: BIRDCLAW_DIGEST_LANGUAGE)",
+		)
+		.option("--refresh", "Bypass the local digest cache")
+		.option("--max-tweets <n>", "Maximum tweet context", "5000")
+		.option("--max-links <n>", "Maximum linked articles", "12")
+		.option("--no-live-sync", "Use only the local database")
+		.option(
+			"--live-mode <mode>",
+			"Live timeline mode: xurl, bird, or auto",
+			"xurl",
+		)
+		.action(async (options) => {
+			await autoUpdateBeforeRead();
+			const digestOptions = buildDigestOptions("today", options);
+			if (!digestOptions) return;
+			await runDigestCli(digestOptions);
+		});
+
+	program
+		.command("digest [period]")
+		.description("Stream an AI digest for today, 24h, yesterday, or week")
+		.option("--account <accountId>", "Account id")
+		.option("--include-dms", "Include private DM context")
+		.option("--since <isoDate>", "Start of explicit window")
+		.option("--until <isoDate>", "End of explicit window")
+		.option("--model <model>", "OpenAI model id")
+		.option(
+			"--language <tag>",
+			"Report language as a Unicode locale id, e.g. zh-CN (env: BIRDCLAW_DIGEST_LANGUAGE)",
+		)
+		.option("--refresh", "Bypass the local digest cache")
+		.option("--max-tweets <n>", "Maximum tweet context", "5000")
+		.option("--max-links <n>", "Maximum linked articles", "12")
+		.option("--no-live-sync", "Use only the local database")
+		.option(
+			"--live-mode <mode>",
+			"Live timeline mode: xurl, bird, or auto",
+			"xurl",
+		)
+		.action(async (period, options) => {
+			await autoUpdateBeforeRead();
+			const digestOptions = buildDigestOptions(period, options);
+			if (!digestOptions) return;
+			await runDigestCli(digestOptions);
+		});
+}

package/src/cli/register-compose.ts

@@ -0,0 +1,40 @@
+import { createDmReply, createPost, createTweetReply } from "#/lib/queries";
+import type { CliCommandContext } from "./command-context";
+
+export function registerComposeCommands({
+	program,
+	print,
+	asJson,
+	autoSyncAfterWrite,
+}: CliCommandContext) {
+	const composeCommand = program
+		.command("compose")
+		.description("Create local/xurl actions");
+
+	composeCommand
+		.command("post <text>")
+		.option("--account <accountId>", "Account id", "acct_primary")
+		.action(async (text, options) => {
+			const result = await createPost(options.account, text);
+			await autoSyncAfterWrite();
+			print(result, asJson());
+		});
+
+	composeCommand
+		.command("reply <tweetId> <text>")
+		.option("--account <accountId>", "Account id", "acct_primary")
+		.action(async (tweetId, text, options) => {
+			const result = await createTweetReply(options.account, tweetId, text);
+			await autoSyncAfterWrite();
+			print(result, asJson());
+		});
+
+	composeCommand
+		.command("dm <conversationId> <text>")
+		.description("Reply inside an existing DM conversation")
+		.action(async (conversationId, text) => {
+			const result = await createDmReply(conversationId, text);
+			await autoSyncAfterWrite();
+			print(result, asJson());
+		});
+}