postplusai/postplus-skills
32 skills20.8k installs608 starsGitHub
Install
npx skills add https://github.com/postplusai/postplus-skillsSkills in this repo
1Audio Transcriptionaudio-transcription is a Generative Media skill from postplusai/postplus-skills in the Media and Creative Production family that transcribes podcast, interview, or voice memo audio through PostPlus hosted Whisper endpoints. The skill accepts remote HTTPS audio URLs or local files after a PostPlus media-file upload, then returns persistent transcript text plus optional timestamps suitable for subtitles, multilingual speech search, and durable request/response records. Developers reach for audio-transcription when the input is audio-only speech-to-text—not video pipelines covered by video-transcription. Outputs include plain transcripts, timed segments, and subtitle-ready artifacts with API persistence for downstream automation.886installs2Video Transcriptionvideo-transcription is a skill from postplusai/postplus-skills in the Media and Creative Production family that transcribes video into timed transcripts and subtitle-ready artifacts. It accepts remote HTTPS video URLs or local files after a PostPlus media-file upload, then runs hosted Whisper video-to-text to extract speech with timestamps. Developers reach for video-transcription when the input is video and the goal is caption generation, multilingual transcripts, edit-prep timing, or speech extraction—not semantic visual analysis, which belongs to the separate video-analysis skill. The skill fits content pipelines, accessibility captioning, and post-production workflows that need machine-readable timed text from spoken audio tracks.796installs3Video Analysisvideo-analysis is a skill in postplusai/postplus-skills that analyzes local or downloaded social video files with the official Gemini API, defaulting to gemini-3.5-flash prompts for TikTok and Reels workflows. It produces shot-beat timelines, hook breakdowns, voiceover or on-screen text capture, and natural Markdown strategy notes linked back to source metadata. The skill sits in the media-production family for media and creative production tasks where title and description metadata are insufficient. Developers invoke it when a user supplies a video file or URL and asks to watch, deconstruct, or analyze creative structure rather than guessing from tags alone. Reach for video-analysis when you need upload-based multimodal review of short-form clips for content iteration, competitive teardowns, or creative briefs grounded in actual frames and audio.791installs4Google Trends Researchgoogle-trends-research is a PostPlus agent skill in the Marketplace, Sourcing, and Growth family for mining Google Trends platform data during early validation. The skill guides topic discovery, keyword momentum tracking, regional interest breakdowns, rising-query watchlists, and search-intent sourcing while treating Google Trends as a search-intent signal—not a proxy for platform content heat or marketplace demand. It applies shared rulebook guidance from postplus-shared so agents follow consistent research workflows instead of ad-hoc trend scraping. Developers reach for google-trends-research when evaluating whether a keyword lane has sustained lift, which geographies drive interest, or which related queries are breaking out before scoping a SaaS feature, landing page, or SEO backlog. Outputs are structured watchlists and momentum notes agents can fold into scope decisions, copy tests, and content roadmaps without over-interpreting trends as purchase intent.770installs5Facebook Researchfacebook-research is a skill in postplusai/postplus-skills that researches public Facebook pages, groups, and posts through the PostPlus Cloud hosted public-content collection surface. It belongs to the platform-research family alongside LinkedIn and YouTube skills and applies shared postplus-shared rulebook and user-guidance rules before any collection run. The skill requires confirming the question can be answered from public Facebook content and refuses tasks that need private or non-public data. Developers use it when they want metrics and summaries grounded in actual posts rather than assumptions about audience behavior or competitor messaging. Outputs are source-linked research briefs suitable for positioning docs, content calendars, or validation memos. Reach for facebook-research when the research question is explicitly about public Facebook surfaces and you already use PostPlus skills for other social platforms.754installs6Postplus Sharedpostplus-shared is the shared rulebook in postplusai/postplus-skills for every PostPlus ad and research skill. It forbids treating ad management as one free-form agent task and splits work into five stages: connect account, read data, normalize, diagnose and propose, then approve and apply. A default mermaid flow routes user intent through ads-router into connect or discover, read or normalize, diagnose, propose patch, a human approval gate, and only then apply changes. Platform-specific skills in the family must load postplus-shared first so spend-affecting writes never run before explicit approval. Developers reach for postplus-shared when wiring agents to Meta, Google, TikTok, or other PostPlus ad surfaces and need consistent safety boundaries across Instagram research, Facebook research, Google Trends, and paid ad diagnostics. Use it before any task that touches ad accounts, campaign reads, or budget-impacting patches.720installs7Youtube Researchyoutube-research is a PostPlus research skill in the platform-research family that queries public YouTube data through PostPlus Cloud instead of private subscriber analytics. Agents call `postplus research scrape youtube-videos` for public video search and `postplus research collect youtube-channel-summary` or comment-collection routes for hosted summaries. The released collection does not expose subscriber identities, so comment authors remain public proxies rather than subscriber bases. Outputs include channel metadata, public video records, comment-sample findings, source URLs, and available metric fields with explicit unavailable-data notes. Developers reach for youtube-research when benchmarking creators, validating positioning, or feeding benchmark-to-brief and video-analysis workflows during early content strategy work.719installs8Social Media Publishersocial-media-publisher is a PostPlus agent skill in the postplus-skills catalog of 41 skills that routes social publishing through PostPlus Cloud instead of third-party credentials. Agents run `postplus publish <operation> --request input.json --output result.json` for create, update, list, delete, schedule, publish, and analytics, applying shared rulebook and approval rules from postplus-shared. Channels onboard via PostPlus invite-link OAuth, and side-effecting operations can require quote confirmation before spend. Developers reach for social-media-publisher when they want agent-driven social scheduling with human approval boundaries, draft-by-default workflows, and hosted workspace identity rather than wiring Buffer, Hootsuite, or native APIs directly in code.712installs9Tiktok Researchtiktok-research is a PostPlus skill in the TikTok family that performs bounded organic research across keywords, hashtags, competitor and creator profiles, comment sampling, trend discovery, and benchmark summaries. It applies shared rulebook guidance from postplus-shared and produces dataset summaries developers can feed into content pipelines, persona docs, or competitive briefs. Use tiktok-research when you need TikTok metadata and audience signals without paid ad intelligence or local audio extraction. Route paid ad questions to tiktok-ad-research and sound-download tasks elsewhere. The skill keeps research scoped so teams validate organic opportunities before committing creative or spend.710installs10Frame Extractionframe-extraction is a PostPlus skill in the Media and Creative Production family that selects useful still frames from local video files based on task intent. It produces contact sheets and review packs for persona research, shot breakdowns, product visibility checks, UI walkthrough captures, visual-style review, and CTA or compliance validation—not broad video summarization. Use frame-extraction when downstream work needs specific frames for ad creative, documentation screenshots, or legal review rather than automated transcript-only analysis. Pair it with video-oriented siblings when full clip analysis is required, but reach for frame-extraction when still selection quality drives the deliverable.708installs11Image Batch Runnerimage-batch-runner is a PostPlus agent skill in the Media and Creative Production family that runs fact-grounded hosted image generation batches for short-form video workflows. Developers invoke it after persona and concept inputs exist to produce persona images, first-frame candidates, and light consistency edits for vertical video. The skill saves raw provider responses and normalized local asset manifests plus prompt records and reusable model-call metadata, staying anchored to benchmark-backed persona locks. Reach for it when the next step is batch image assets rather than defining personas or writing scripts from scratch.708installs12Editing Decision Engineediting-decision-engine is a PostPlus agent skill from postplusai/postplus-skills in the Media and Creative Production family. It plans short-form post-edit decisions from A-roll, B-roll, scripts, and reference videos—deciding where to stay on face, where to insert proof B-roll, and how to apply reference patterns—then packages an actionable edit plan for human editors or downstream timeline tools. Developers and producers reach for editing-decision-engine when they need structured cut guidance for social video, not generic scene detection or final render pipelines. The skill deliberately avoids rendering; it delivers beat-by-beat editorial decisions agents or editors can execute in NLE or automated timeline workflows.704installs13Sourcing Selectionsourcing-selection is a skill from postplusai/postplus-skills in the Marketplace, Sourcing, and Growth family (familyId marketplace-sourcing). It combines supply-side and demand-side evidence into a practical sourcing judgment for a product, niche, or shortlist without collapsing platform data collection and business judgment into one step. Use when a product idea, supplier signal, channel opportunity, or shortlist needs a cautious go/no-go or next-step memo before inventory or supplier commitments. Do not use when raw platform scraping alone is sufficient or when final purchase orders should be placed without human review. The skill separates evidence synthesis from execution, producing decision-ready memos for ecommerce operators validating marketplace opportunities.698installs14Subtitle Packagersubtitle-packager is a PostPlus skill in the Media and Creative Production family that packages an already-normalized timed transcript JSON into subtitle deliverables. It handles subtitle chunking, timing normalization, and export to SRT and ASS formats plus editor-ready caption bundles. The skill explicitly does not call speech-to-text or video analysis models, so it assumes stable transcript input from an upstream step. Developers and media engineers reach for subtitle-packager when caption timing and format conversion are the remaining work after transcription. It fits video pipelines, localization handoffs, and editor workflows where re-transcribing would waste compute and risk timing drift.697installs15Reference Contract Builderreference-contract-builder is a Generative Media skill from postplusai/postplus-skills in the Routing & Contracts family that builds explicit learn/do-not-copy contracts for generative prompts. The skill activates when prompts use benchmark videos, contact sheets, frames, product images, persona images, style boards, or audio references and need clear rules on what the model should learn versus which identity elements must change. Developers reach for reference-contract-builder to exclude certain references from first tests and document exactly which visual traits are transferable. It prevents benchmark likeness from contaminating initial generation runs in PostPlus image and video pipelines.695installs16Social Media Extractorsocial-media-extractor is a high-level entry skill from postplusai/postplus-skills in the Routing and Contracts family for public Instagram and Facebook/Meta social data extraction. When a developer needs ad creative references, competitive benchmarks, creator shortlists, or UGC signals but has not named a platform-specific workflow, this skill routes the request to the correct downstream skill—Instagram benchmark or creator work to Instagram skills, and Facebook/Meta page, group, or post evidence to facebook-research. Developers reach for social-media-extractor at the start of competitive or creative research sessions to avoid picking the wrong specialized extractor. The skill defines scope boundaries and handoff contracts before platform-specific extraction begins.694installs17Benchmark To BriefBenchmark To Brief is a PostPlus agent skill for solo builders and small creative teams who already finished benchmark research on short-form video and need an execution-ready strategy layer. It ingests validated evidence—reports, master tables, pattern tables, comment analyses, performance findings—and outputs campaign briefs, concept candidates, hook options, lane priorities, and test plans that stay anchored to source facts. The skill refuses greenfield invention: unsupported personas, angles, or claims are out of scope. It stops at script-ready briefs and routes final scripting to human review and asset generation to downstream workflows. Use it as the strategy bridge between research tables and production, not as a brainstorming substitute when no data exists.692installs18Voice Batch Runnervoice-batch-runner is a PostPlus media-production skill for persona-aware voice asset generation in short-form video workflows. After persona registries and scripts exist, the skill drives hosted voice design, script-specific audio takes, and reusable voice identities while preserving continuity across many videos. Developers reach for voice-batch-runner when they need local audio assets, reviewable voice iterations, and manifest files without manually re-prompting or losing persona tone between clips. The skill belongs to the Media and Creative Production family and focuses on polling, versioning, and manifest management rather than script writing or final edit assembly.691installs19Prompt Preflight Qaprompt-preflight-qa is a PostPlus Routing & Contracts skill that runs a credit-saving QA pass on existing storyboard or full video-request prompts before generation. It flags weak first frames, drift risk, missing constraints, bad product timing, and generic ad-like language, then outputs a concise readiness report. Developers and creative engineers reach for prompt-preflight-qa when a prompt draft already exists and model spend should not fire on incomplete hooks or ambiguous shot constraints. The skill sits in the postplus-shared guidance family and focuses on pre-generation review rather than writing prompts from scratch.690installs20Video Request Architectvideo-request-architect is a PostPlus skill in the Routing and Contracts family that converts approved scene logic, storyboards, beat sheets, or prompt plans into self-contained short-form video generation requests. The architecture is provider-agnostic so segments can later map cleanly into Seedance or other video models without rewriting creative structure. Developers use it when a long script must be split into independent generation segments with stable beat boundaries already signed off. The skill does not invent story content; it structures approved beats into request objects ready for downstream model routing. Media and AI pipeline engineers reach for video-request-architect at the handoff between creative approval and model-specific prompt execution in automated video production workflows.689installs21Reference Decodereference-decode is a PostPlus routing and contracts family skill that decodes benchmark videos, contact sheets, frame sets, or rough creative notes into reusable prompt structure before storyboard or video generation work begins. The skill extracts hook essence, the viewer question, must-copy visual grammar, and forbidden drift constraints so downstream prompts stay faithful to the reference without vague creative briefs. Developers and creative technologists reach for reference-decode when a benchmark clip or mood board must become explicit generation instructions rather than subjective descriptions. The workflow supports both reference-backed briefs and no-reference cases that still need disciplined prompt scaffolding. reference-decode outputs structured objects consumed by storyboard writers and video model prompt authors in PostPlus pipelines. The skill does not render video, run generation jobs, or replace final copywriting; it front-loads decode artifacts that keep hooks and visual rules consistent across generative steps.688installs22Video Batch Runnervideo-batch-runner is a PostPlus Media and Creative Production skill that turns approved image, script, voice, or prompt-plan inputs into hosted talking-head, Seedance, or reference-motion video renders. It generates local render manifests, downloads finished clips, and maintains replaceable routes with pollable checkpoints so persona and concept continuity survive across a batch. Developers use video-batch-runner when upstream creative assets are signed off and the next step is scalable short-form video generation without manually polling each InfiniteTalk or Seedance 2.0 job.688installs23Creative Qacreative-qa is an agent skill from postplusai/postplus-skills in the Media and Creative Production family that records human-in-the-loop quality judgments for generated images, voice takes, video renders, and final exports. After a person reviews an asset, the skill captures structured verdicts, reasons, issue categories, blame stage attribution, and rerun guidance instead of leaving subjective approval as untracked chat history. Developers reach for creative-qa in short-form and generative media workflows where QA must be auditable—knowing which pipeline stage failed, why an image or voice take was rejected, and what to regenerate next. The skill fits post-generation review gates before publishing or downstream edit assembly. It complements PostPlus generation skills by focusing on judgment capture and remediation notes rather than model selection or rendering commands themselves.687installs24Media Routermedia-router is an agent skill from postplusai/postplus-skills in the Routing and Contracts family that decides which PostPlus media workflow should run before any tokens are spent on the wrong model chain. When a developer supplies audio, video, transcript, subtitle, A-roll, or B-roll material, media-router determines whether to transcribe, package subtitles, add semantic understanding, or build an edit-ready plan based on the output quality target. Developers reach for media-router at the first decision point in short-form production pipelines—before transcription, beat mapping, or B-roll planning skills execute independently and potentially duplicate work. The skill applies shared postplus-shared rulebook guidance so routing stays consistent across PostPlus media skills. Use it whenever the user's request could map to multiple media-understanding paths and the priority is picking the right downstream skill rather than running a default transcription pass.687installs25Seedance Submitterseedance-submitter is a skill from postplusai/postplus-skills in the Media and Creative Production family for Seedance 2.0 video generation routed through PostPlus Cloud. The workflow covers splitting UGC scripts into render segments, uploading product or storyboard reference images, building and validating promptPlan request JSON, submitting jobs, polling prediction status, and handing off local render paths. Developers reach for seedance-submitter when automating short-form video creation from product shots, voiceover copy, or scripted UGC content without manually driving the PostPlus dashboard. The skill fits content pipelines that batch video renders and need repeatable job submission, status polling, and debug steps for failed predictions.679installs26Slideshow Producerslideshow-producer is a PostPlus media-production skill in the postplusai/postplus-skills repository that orchestrates TikTok slideshow and Instagram carousel creation from vibe-driven prompts through export-ready assets. The workflow saves a local JSON slide manifest as routing source of truth, routes slides to image-gpt-image-2-text or image-gpt-image-2-edit based on reference images, and runs batch generation through the postplus media CLI with aspect ratios such as 9:16 for TikTok and 4:5 for Instagram carousels. Developers reach for slideshow-producer when they need script-aligned slide planning, reviewable manifests, GPT Image 2 batch generation, and composited text overlays without calling provider APIs directly. The skill integrates with postplus doctor, postplus media schema, and quote-confirmation flows, then hands off final image paths and updated manifests for creative QA or publishing skills.526installs27Storyboard Grid WriterStoryboard Grid Writer is a PostPlus media-production skill for solo builders and small creative teams who use agents to plan short-form visuals before any provider call. Install it when your generation controller needs a contact sheet or 6–9 panel beat plan with hook mechanics translated into frame-by-frame action, product timing, and drift guardrails. The skill enforces a simple split: keep the shared prompt short and put scene logic in detailed panel lines, which reduces rework when image or video architectures consume the pack. It expects structured inputs such as creative format, forbidden drift, and must-copy visual grammar, and it refuses plain script-only or execution-submission jobs where reference contracts are still unresolved. Best used after reference meaning is settled (or explicitly marked no-reference) and before final runner approval.387installs28Image GenerationImage Generation is a PostPlus routing-contract skill that acts as the controller for any agent session where the deliverable is still or batched images. Solo builders and small teams use it when a user asks for text-to-image, edits, persona shots, product renders, banners, thumbnails, or storyboard panels, but the runtime must not fire jobs until inputs are understood and the correct runner payload exists. The skill walks a four-step gate: detect input type, assign a task class, apply model and reference rules, then emit a structured handoff for image-batch-runner. That separation keeps expensive API calls and batch orchestration in one executor while this skill stays deterministic and auditable. It fits Claude Code, Cursor, and Codex flows that chain PostPlus family skills for creative pipelines without mixing video, audio, or edit-plan controllers into the same step.378installs29Video GenerationVideo Generation is a PostPlus routing-contract skill that acts as the controller layer for agent-driven video creation. Solo and indie builders use it when a user asks for a clip, storyboard reel, UGC piece, podcast visual, reference-based video, talking-head, image-to-video, text-to-video, or research-handoff video and the agent must decide what kind of work this is before anyone hits a provider API. The skill reads the natural-language intent, maps it to a task class, and determines whether references or a storyboard must be resolved upstream. It deliberately does not submit jobs; that keeps failures and cost surprises out of casual chat turns. After classification, it routes approved structure to video-request-architect and execution-ready payloads to seedance-submitter or video-batch-runner depending on batch versus single-shot needs. If the user only wants transcription, subtitles, semantic media analysis, or edit decisioning, media-router is the correct entry point instead. If architecture is already signed off, builders should bypass this controller and call a runner directly. The pattern fits content creators and product teams who wire Claude Code or Cursor into generative378installs30Generation Routergeneration-router is a PostPlus routing-contract skill that sits at the front of generative media agent workflows. When a solo builder asks to generate, edit, dub, animate, or assemble image, video, audio, or analysis-derived assets, the skill forces a classification step: image vs video vs audio (and related UGC flows) before any provider or model is touched. That prevents expensive wrong-path runs and keeps transcription-first or analysis-first jobs on media-router instead. It is intermediate complexity because you need to understand PostPlus controller names and when inputs are already sufficient to bypass routing. Use during creative automation builds and launch asset batches; skip it for pure copywriting, publishing, or research-only tasks as the skill itself states.377installs31Audio GenerationAudio Generation is a PostPlus routing-contract skill for solo builders and small teams automating spoken and dubbed media with agents. Before any TTS, persona voice, voice change, translated dub, cloned take, podcast audio, or lip-sync handoff runs, the skill decides the audio task class, reference policy, and which downstream runner owns execution. It prevents runners from guessing voice design, translation rules, or lip-sync intent. Use it when the desired outcome is generated audio or audio staged for a video render; skip it when the user only needs transcription or subtitles (media-router), when requests are already normalized for voice-batch-runner, or when the primary goal is a full video pipeline without a clear audio class. The skill fits agent-tooling and content workflows where predictable contracts matter more than one-off chat improvisation.376installs32Ugc Flowugc-flow is a PostPlus workflow skill that orchestrates a full UGC-style creative production line for solo builders and small teams promoting products without hiring a full studio. It starts from product or research evidence, defines creator persona and tone, builds a storyboard, then hands off to dedicated generation runners for images, video clips, audio, subtitles, and montage planning—with explicit QA gates so invented claims do not slip into ads. The skill is a coordinator: it sequences public skills and checkpoints rather than calling providers itself. Use it when the user wants a standard line instead of one isolated render, spanning Launch distribution work and Grow lifecycle content refreshes. It pairs naturally with PostPlus media skills when requests are provider-ready.375installs