Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Source and send relevant reaction GIFs in WhatsApp chats using safe filters and deterministic ranking. Use when the user asks for a GIF/meme/reaction, or whe...
Source and send relevant reaction GIFs in WhatsApp chats using safe filters and deterministic ranking. Use when the user asks for a GIF/meme/reaction, or whe...
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
Source safe, context-matching GIFs and send exactly one high-quality result to WhatsApp.
Decide if GIF is appropriate. Build concise search query from user intent (2-5 words). Run scripts/find_gif.py to fetch and rank candidates. Send top result with message tool to WhatsApp (prefer local-file payload mode). If send fails, retry once with next candidate. If still failing or no results, send concise fallback text.
Prefer GIF when: The user explicitly asks for GIF/meme/reaction. A brief emotional signal is enough. Tone benefits from a visual response. Avoid GIF when: Topic is serious/sensitive (medical, legal, conflict, grief). Group context is formal or unclear. User asked for text-only response.
Use these deterministic helpers: python3 skills/openclaw-whatsapp-gif/scripts/find_gif.py "congrats celebration" --limit 5 --json python3 skills/openclaw-whatsapp-gif/scripts/find_gif.py "great job" --limit 3 --json --target "+1234567890" python3 skills/openclaw-whatsapp-gif/scripts/send_gif.py "+1234567890" "great job celebration" --json python3 skills/openclaw-whatsapp-gif/scripts/send_gif.py "+1234567890" "great job celebration" --delivery-mode local --json python3 skills/openclaw-whatsapp-gif/scripts/send_gif.py "+1234567890" "great job celebration" --payload-only Behavior: Reads TENOR_API_KEY and/or GIPHY_API_KEY. Applies safe filtering (contentfilter=low / rating=g + blocklist). Uses retry + backoff for network fetches. Ranks by keyword overlap + lightweight preference. If no API keys or no provider results, uses built-in safe fallback catalog. Optional web scraping fallback is available only when allowWebScrapeFallback is enabled in references/policy.json. Expands intent queries using common reaction aliases. Deduplicates repeated URLs across providers. Returns URL only (default) or candidate list (--json). With --target, prints a ready-to-send WhatsApp message payload in JSON. With send_gif.py --payload-only, prints only the final message payload for direct tool handoff. send_gif.py validates media size/type, retries next candidates when a URL is bad, and writes to OS temp cache. Telemetry logging is disabled by default and can be enabled by policy or CLI flag. Remote URL fallback is disabled by default and must be explicitly enabled in policy. Runtime policy is configurable in references/policy.json.
Use message with: action: "send" channel: "whatsapp" target: <chat id/recipient> media: <gif/mp4 url> caption: <optional short line> gifPlayback: true If provider rejects .gif, retry using next candidate URL (prefer MP4).
If no suitable GIF or delivery fails twice: Send concise text preserving intent. Do not keep retrying in a loop.
Provider notes: references/providers.md
Messaging, meetings, inboxes, CRM, and teammate communication surfaces.
Largest current source with strong distribution and engagement signals.