pagepop-skill

Create and continue PagePop content-generation conversations from a host app. Use when the user wants PagePop to generate, refine, or retrieve content artifacts such as social posts, slides, documents, images, or design drafts.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "pagepop-skill" with this command: npx skills add jujian-pp/pagepop-skill

PagePop Skill

This skill connects a host application to PagePop.

On first use, the user opens a PagePop authorization page in a browser and confirms access. After authorization, the skill stores a local access key and reuses it for later requests.

Capabilities

  • Start a PagePop generation request.
  • Continue the current PagePop conversation by default.
  • Start a new conversation when explicitly requested.
  • List locally saved conversations for switch-chat flows.
  • Stream PagePop progress and artifact events.
  • Emit a generic artifact delivery view plus channel-aware presentation payloads for host applications.
  • Check whether the installed package should be updated.

Commands

python3 scripts/pagepop_skill.py status
python3 scripts/pagepop_skill.py auth
python3 scripts/pagepop_skill.py conversations
python3 scripts/pagepop_skill.py stream --goal "Create a product launch deck"
python3 scripts/pagepop_skill.py stream --new-chat --goal "Create a rednote post about camping gear"
python3 scripts/pagepop_skill.py resume-stream --conversation-id conv_xxx --offset 0

stream --goal always submits a chat request before reading SSE events. Use resume-stream when the host only needs to replay or continue reading events for an existing conversation without creating a new task. When --offset is omitted, resume-stream uses the saved SSE cursor for that conversation.

Configuration

Production packages are built with a generated skill-manifest.json.

Common environment variables:

  • PAGEPOP_API_BASE_URL
  • PAGEPOP_WEB_BASE_URL
  • PAGEPOP_SKILL_ID
  • PAGEPOP_SKILL_STATE_DIR
  • PAGEPOP_SKILL_REGION
  • PAGEPOP_SKILL_SOURCE_APP
  • PAGEPOP_SKILL_DISPLAY_APP_NAME
  • PAGEPOP_SKILL_RETURN_MODE
  • PAGEPOP_SKILL_RETURN_TARGET
  • PAGEPOP_SKILL_UPDATE_CHANNEL
  • PAGEPOP_SKILL_WAIT_FOR_AUTHORIZATION
  • PAGEPOP_SKILL_ARTIFACT_DIR
  • PAGEPOP_SKILL_DOWNLOAD_IMAGES

The legacy PAGEPOP_OPENCLAW_* names are still accepted as compatibility aliases for existing OpenClaw installations. New integrations should use the PAGEPOP_SKILL_* names. Host-specific usage notes belong in adapters/.

Before invoking this skill, host applications should determine whether the current user is in mainland China or outside mainland China. If PAGEPOP_API_BASE_URL is not set, production domain selection is region-based: mainland China (PAGEPOP_SKILL_REGION=CN or an Asia/Shanghai-style mainland timezone) uses https://pc-api.pagepop.cn; non-mainland users use https://pc-api.pagepop.ai. Missing region data defaults to the global .ai domain and emits an integration_warning.

Host Integration

When a host invokes this skill on behalf of another app, it should pass launch context before authorization:

  • source_app
  • display_app_name
  • return_mode
  • return_target
  • region or PAGEPOP_SKILL_REGION, after deciding mainland China vs non-mainland

If no launch context is provided, the authorization page falls back to the default host label.

Artifact delivery events include:

  • presentation: channel-neutral content for fallback renderers.
  • channel_presentations.slack: Slack Block Kit blocks.
  • channel_presentations.feishu: Feishu interactive-card payload plus media upload hints.
  • target.preferred_channel: normalized from source_app when the host passes launch context.

Hosts should prefer their matching channel_presentations entry, then fall back to presentation.fallback_text. Feishu hosts should avoid rendering raw URLs from generic presentation; use the Feishu card buttons or Feishu-specific fallback text so underscores in URLs are not truncated by lark_md auto-linking. When composing a Feishu chat message manually, send channel_presentations.feishu.media.local_image_messages as image messages first, then send channel_presentations.feishu.plain_text or artifact.display_text; do not copy raw image URLs from artifact.text or sse_event.data into lark_md.

Output

The skill emits JSON Lines. Important event kinds include:

  • auth_required
  • auth_authorized
  • chat_context
  • conversation_history
  • chat_submitted
  • stream_resumed
  • progress_update
  • artifact_update
  • artifact_ready
  • artifact_delivery
  • sse_event
  • stream_finished
  • skill_update_available
  • skill_update_required
  • error

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

Eversince

Eversince is a creative agent that plans and executes across image, video, audio, and motion graphics. It lives in a purpose-built environment for creative w...

Registry SourceRecently Updated
1590Profile unavailable
General

Insta Content Engine

Find trending topics, create editorial-style social media graphics, and post to X/Twitter and Instagram. Includes image generation with photographic backgrou...

Registry SourceRecently Updated
1560Profile unavailable
Automation

AutoClaw Browser Automation

Complete browser automation skill with MCP protocol support and Chrome extension

Registry SourceRecently Updated
1.2K0Profile unavailable
Automation

Lemnos Cost Guard

Real-time API cost tracking, context bloat detection, and budget enforcement for OpenClaw agents. Use when setting up cost guardrails, checking daily spend,...

Registry SourceRecently Updated
5940Profile unavailable