Agent Skills

youtube-research

Research public YouTube channels, video metrics, and comment samples; obtain downloadable video records when requested. Use for content and audience evidence.

Install

npx skills add https://github.com/postplusai/postplus-skills --skill youtube-research
SKILL.md

YouTube Research

Use this skill for public YouTube channel summaries, audience comment samples, downloadable video records, and public video metrics through PostPlus.

Before Collection Boundary

The released collection does not expose subscriber identities. Use channel metadata and comments as public proxies, and do not present comment authors as the subscriber base.

Research Routes

Use youtube-videos for public video records, youtube-channel-summary for channel facts, youtube-comments for audience language, and youtube-video-download only when downloadable records are explicitly needed. PostPlus handles execution details.

Evidence Quality

  1. Check channel facts, video records, and comments against the question; comment authors are not the subscriber population.
  2. If a completed pass misses, refine the channel/video source or supported query rather than treating snippets as comments.
  3. Allow at most two changed follow-up passes after successful but insufficient results, within approved scope and budget; do not repeat an identical request or hide a failed/pending operation.
  4. Stop when sufficient, at the bound, or when another pass would not help. Report useful evidence and uncertainty; preserve raw results and source links.

Public video metrics are snapshots; preserve source URLs and dates, and do not infer subscriber identities. Full machine fields belong to postplus research schema --route <route> --json; consult it only when required for processing, not before every request.

postplus research run youtube-videos \
  --url "https://example.com/source" \
  --wait \
  --output ./result.json
postplus research run youtube-channel-summary \
  --channel "@example" \
  --wait \
  --output ./result.json

Follow the CLI's structured result and reported next action; do not infer recovery from free-text messages. Wait for explicit user approval when requested; an action does not authorize spending, publishing, or overwriting. Resume the same operation through its returned checkpoint or action; never resubmit uncertain work, repeat exhausted recovery, or switch providers to bypass failure.

Default Workflow

  1. For channel research, collect a channel summary.
  2. For audience research, collect a small comments sample.
  3. For broad public video discovery, compile a small public-video plan and run the youtube-videos route.
  4. If a run is pending, preserve its result file and resume with postplus research run --resume-from <result.json>.
  5. Keep observation separate from inference, especially for audience claims.

While a run is pending, tell the user the research is continuing from a saved checkpoint and continue independent brief or source-review work.

Output

Return channel metadata, public video records, comment-sample findings, source URLs, metric fields that were present, and clear notes about unavailable subscriber identities.

Failure Modes

  • Stop if the request includes non-YouTube platforms.
  • Stop if no public YouTube URL, channel, or query can be used.
  • Stop if PostPlus cannot resolve the channel handle or URL.
  • Stop on PostPlus service, auth, DNS, proxy, network, or malformed-output hard errors.
  • Do not promise private audience identities or logged-in analytics.

Handoff

Benchmark findings can feed benchmark-to-brief, media-analysis, or broader cross-platform synthesis.

Public Command Boundary

  • Choose the smallest matching command or workflow from the user input and run it directly.

  • Readiness diagnostics: postplus doctor --skill youtube-research.

  • Inspect a route with postplus research run <route> --help only when needed.

  • Run postplus research run <route> --<url/channel flags> --limit <n> --wait --output <result.json>.

  • Use only the semantic flags shown by the selected route.

  • Keep the first pass bounded; expand only after inspecting the first result.

  • If the CLI returns a quote-confirmation challenge, obtain user approval for its scope and cost before running postplus quote confirm --json --challenge-file <challenge.json> and retry with the returned token.

Related skills

researchmattpocock575KInvestigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.paper-context-resolverlllllllama451KRigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing Renv-and-assets-bootstraplllllllama450KRigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.ai-research-explorelllllllama311KRigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, fair comparison, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding, narrow c

Search skills and MCP servers

Fuzzy search across 23,137 skills and servers