version unknown

skill-last30days-openclaw

@weekendli

# last30days-openclaw

Choose how to use this skill

You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.

1. Native installer

This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.

Do not guess an installer command or replace an existing version without reviewing the diff.

2. Complete package recommended

Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.

No complete ProSkills package is published for this listing yet.

3. Prompt-only

Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.

Need only the instruction file?

Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.

No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.

7.5/10
Verified
Apr 29, 2026

// RATINGS

⭐GitHub Stars
⭐ 0 on GitHubGitHub ↗

New / niche

🟢ProSkills ScoreAI Verified
7.5/10
📍

Not yet listed on ClawHub or SkillsMP

// README

last30days-openclaw

OpenClaw adaptation of @mvanhorn's last30days-skill: https://github.com/mvanhorn/last30days-skill

Attribution (required)

This repository is a separate adaptation for OpenClaw packaging and runtime conventions. It does not replace the upstream project and should not be represented as the original work.

What came from upstream vs what changed

Upstream/original work (credited to @mvanhorn)

  • Research engine (scripts/last30days.py)
  • Source connectors and scoring pipeline (scripts/lib/*)
  • Watchlist/briefing/history architecture (scripts/watchlist.py, scripts/briefing.py, scripts/store.py)
  • X search vendor integration (scripts/lib/vendor/bird-search/*)
  • Research methodology and source synthesis structure

OpenClaw adaptation work (this repo)

  • OpenClaw skill metadata (skill.json)
  • OpenClaw usage docs (SKILL.md)
  • OpenClaw workspace path defaults:
    • ~/.openclaw/workspace/.secrets/last30days.env
    • ~/.openclaw/workspace/data/last30days/*
  • Setup helper for OpenClaw secrets (scripts/setup_openclaw_env.sh)
  • OpenClaw cron/watchlist runner (scripts/openclaw_watchlist_run.sh)
  • Updated path/help text for OpenClaw operations

Sources supported

  • Reddit
  • X / Twitter
  • YouTube
  • TikTok
  • Instagram Reels
  • Hacker News
  • Polymarket
  • Web search backends (Brave/Parallel/OpenRouter or OpenClaw web_search)

OpenClaw quick start

cd ~/.openclaw/workspace/skills/last30days-openclaw
./scripts/setup_openclaw_env.sh
python3 scripts/last30days.py --diagnose
python3 scripts/openclaw_run.py "openclaw agent frameworks" --quick --search hn,polymarket

API key handling in OpenClaw

Primary secrets file:

~/.openclaw/workspace/.secrets/last30days.env

Recommended keys:

  • SCRAPECREATORS_API_KEY (Reddit + TikTok + Instagram)
  • XAI_API_KEY (optional fallback for X when Bird cookie auth is unavailable)
  • BRAVE_API_KEY / PARALLEL_API_KEY / OPENROUTER_API_KEY (optional native web search)

macOS X cookie support

The vendored Bird module supports browser cookie auth on macOS.

node scripts/lib/vendor/bird-search/bird-search.mjs --whoami

If unavailable, use AUTH_TOKEN and CT0 in the secrets file.

Watchlist + OpenClaw cron integration

Use:

~/.openclaw/workspace/skills/last30days-openclaw/scripts/openclaw_watchlist_run.sh

as the scheduled command in your OpenClaw cron/scheduler job.

Compatibility note

Upstream last30days-skill remains compatible with Claude Code, Pi, and Codex variants. This adaptation is specifically tuned for OpenClaw workspace conventions and tool orchestration.

License

MIT (same as upstream). See LICENSE.


If you publish this adaptation (GitHub/ClawHub), keep the attribution block above intact and link back to the upstream project.