generator-extension extends short video clips to longer formats via ai-powered cloud rendering. users upload clips, describe desired output, and receive 1080p exports within 1-2 minutes using a session-based api with token authentication and credit tracking.

ai-subtitle-converter automates speech-to-text embedding in video files up to 500mb. cloud gpu backend processes mp4, mov, avi, webm in 30-90 seconds, returning 1080p output with language-specific subtitle tracks.

ui-visual-validator codifies a skeptical visual testing methodology emphasizing evidence-based assessment, accessibility compliance, and systematic cross-platform validation. role-based expert for design system and regression testing verification.

cli-anything-openclaw bridges the upstream CLI-Anything methodology into OpenClaw workflows, supporting analysis, design, refinement, and validation of agent-friendly CLI harnesses for GUI applications and repositories.

still-image-to-video converts static images to animated mp4 clips via remote gpu rendering. users upload photos, describe desired motion, and receive 1080p output in 20-90 seconds. no local installation required.

converts arxiv papers from english html to locally-readable chinese static webpages, preserving figures, assets, and css while integrating translated markdown output.

momentic-result-classification uses Momentic MCP tools to analyze failing test runs, categorize failures into predefined buckets, and determine recoverability (recoverable, one-time recoverable, non-recoverable).

bilibili-transcriber automates extraction of bilibili video transcripts and structured summaries via dual-engine architecture (aliyun paraformer cloud or faster-whisper local). accepts bv codes and full urls, handles subtitle detection, audio extraction, and markdown output with timestamps.

caption-video automates subtitle generation for video clips up to 500MB via remote gpu processing. users upload files and describe desired output (language, format) through natural language, receiving 1080p mp4 downloads within 30-90 seconds. built around nemo video api with session-based rendering and credit tracking.

approve-pairing handles manual approval of pending dm pairing requests by directly manipulating openclaw credential files when cli tooling is unavailable, supporting multiple chat platforms.

editor-online-generator automates video editing via natural language, uploading clips and requesting edits (trim pauses, add transitions) that render server-side and return mp4 files in 30-90 seconds.

lightningprox routes inference requests to 5 ai providers (anthropic, openai, together, mistral, google) via bitcoin lightning micropayments. no api keys or accounts required. supports l402 protocol and spend tokens.

editor-anup-sagar automates video editing via cloud gpu, accepting raw footage up to 500mb and returning polished 1080p mp4 exports within 90 seconds. requires nemo token authentication and handles multi-track timelines with text overlays and transitions.

wechat-group-operator automates recurring group posts (morning questions, afternoon followups, evening cases) on windows desktop wechat via python scripts and json content pools, with dry-run capability and cron integration.

consortiumai-create-account wraps a custodial wallet provisioning api, accepting a wallet address and returning account metadata including id, generated custodial address, and timestamps.

video-editing-apps automates trimming, transitions, and mp4 export via remote gpu rendering. users describe edits in natural language, upload clips up to 500mb, and receive polished output in 1-2 minutes without learning professional software.

troubleshooting-dbt-job-errors guides systematic diagnosis of dbt cloud job failures through log analysis, api querying, and data investigation. covers when to use and the principle of investigating root causes before modifying tests.

influencer-video-skill automates raw video editing for creators via cloud gpu rendering. users upload footage, describe desired edits in plain language, and receive 1080p mp4 exports optimized for tiktok and instagram within 1-2 minutes.

google-calendar-update-summary wraps the gog cli to modify event titles only, with jidoka-style validation that retries failed updates up to three times and confirms summary field matches the request.

equity-research-analyst dispatches to nine institutional workflows (initiation, earnings analysis, preview, catalyst calendar, morning note, sector overview, thesis tracking, model updates, idea generation) with structured data sourcing from tradingviewapi and web search fallback.

baidu-pan-per-auth-skill handles oauth2 authorization code flow for baidu pan, covering initial token acquisition, 30-day refresh cycles, and token validation via curl or python scripts.

video-make processes raw footage into 1080p mp4 files via cloud gpu rendering. supports mp4, mov, avi, webm inputs up to 500mb, completes in 1-2 minutes, requires nemo token authentication and session management.

pilot-code-review-pipeline-setup deploys a three-agent code review system (scanner, reviewer, reporter) that chains static analysis through filtering to github/slack notifications. covers role assignment, skill installation, hostname setup, manifest configuration, and inter-agent handshaking via pilotctl.

lattice-protocol provides a decentralized social layer for ai agents with did identity, experience-based reputation, feed curation, and cryptographic attestations. includes cron job templates for autonomous engagement within rate limits.

test-before provides guardrails for agentic wallet operations via caw cli, emphasizing owner approval gates, balance checks, and explicit parameter validation before on-chain execution.

uapp-outlier queries umeng+ anomaly detection across three endpoints: app outlier reports by date, yesterday's anomalies across all apps, and intelligent inspection summaries. credential resolution via user input, config file, or environment variables. returns detailed outlier metadata including type, category, and report urls.

pitch-deck-visuals provides a 12-slide framework with design rules, chart recommendations, and bash/python examples for generating investor pitch decks. covers typography, layout, competitive positioning, and common mistakes but lacks error handling and recovery procedures.

zhoubaoshou transforms scattered work notes into formatted weekly reports by classifying entries into completion, plans, and blockers, then standardizing language to enterprise tone. targets users managing ad-hoc work logs and informal documentation.

algernon-sprint runs timed study sessions with shuffled cards from multiple materials, measuring retention before and after a post-sprint retrieval test to quantify learning gain.

editor-no-generator wraps a cloud video editing backend, accepting natural language descriptions of edits and returning processed mp4/mov files. users upload clips, describe desired cuts and transitions, and retrieve results within 1-2 minutes.

job-search-tracker consolidates application pipeline management, email and linkedin scanning, interview prep materials, and salary negotiation support into a markdown-backed tracker. handles cover letter and follow-up drafting with tone feedback and proactive reminders for stale applications.

comsol-opto-simulation automates multiphysics optical, semiconductor, and thermal modeling via python/mph. requires user-supplied geometry, materials, boundary conditions, and solver configuration before execution.

motion-planning wraps Motion task and project management through ClawLink, a credential abstraction layer. covers inspection, creation, and updates of tasks, projects, and schedules via dynamically discovered tools.

coordinator-agent monitors an openclaw agent fleet once daily, delivering a priority-sorted briefing to telegram/discord/slack only when changes occur. includes trend detection, self-healing cron retriggering, and silent mode for zero-change days.

talent-profile-generator produces role-specific competency models and hiring criteria by synthesizing enterprise context (culture, level, strategy) with public job market data from chinese recruitment platforms. outputs structured talent portraits, interview frameworks, and candidate personas tied to business outcomes.

ikuai-report generates a 5-tab html traffic dashboard from ikuai-cli json data using chart.js animations. requires router ip and api token configuration, outputs to /tmp/ikuai-report.html.

text-to-video-chinese-ai converts chinese text prompts (txt, docx, pdf, srt) into 1080p mp4 videos via cloud gpu rendering, handling auth, session management, uploads, and exports through a nemo backend api with token-based credit tracking.

amber-phone-capabilities integrates twilio and openai realtime into openclaw agents, enabling inbound screening, outbound calls, and calendar scheduling via a setup wizard and mcp plugin. voice workflows are confirmation-gated and operator-reviewed.

publish-wechat-official-account-article-infinite guides authors through wechat official account article publishing for the "infinite" publication, with prescriptive sequencing, layout documentation, and checkbox state validation rules.

bizyair-video wraps five video generation modes (image-to-video and keyframe-to-video variants) via bizyair's async api, returning task ids for polling. supports custom dimensions and frame counts with modular model selection.

browser-gemini-search uses openclaw's browser control to navigate chrome and execute searches on gemini.google.com, requiring remote debugging enabled and an active user profile. returns the gemini response text to the user.

accessibility-scanner audits web applications against wcag 2.2 standards, identifying issues in alt text, keyboard navigation, color contrast, aria usage, and semantic html through static analysis and component-level review.

pdfagent enables self-hosted pdf operations (merge, split, compress, convert, ocr) with metered usage output, running via cli without external service dependencies.

whatsapp-otp dispatches one-time passwords via cmi omnichannel rcs using pre-configured templates. handles credential collection, phone number formatting, and api integration with explicit ssl/proxy workarounds documented.

insurance-ai-news aggregates daily ai developments across 13 sectors (9 regions, investment, regulation, key companies, talent) in global insurance, formatted in dual-language takeaway summaries with source links, synced to feishu docs at 08:30 utc.

opencc converts chinese text between simplified, traditional, and regional variants (taiwan, hong kong, japan) via configuration-driven conversion modes. handles phrase-level and character-level transformations with regional awareness.

allclaw enables registration of ai agents on a competitive trading platform where users buy and sell agent shares via hip currency, track elo ratings, and participate in code duels and market predictions.