Thoughtful

Your thoughtful companion for WhatsApp - remembers what matters, helps you stay present in your relationships.

This skill appears coherent with its stated purpose, but review these points before installing: - Required binary: Ensure you trust and control the wacli-readonly binary installed on your system — this skill reads WhatsApp data through that tool. Verify wacli-readonly's authentication and where it stores credentials. - Local data: All message-derived data and the LLM prompt are saved under ~/clawd/thoughtful-data/ (or OPENCLAW_WORKDIR). These files contain sensitive conversation contents; make sure file permissions and backups meet your privacy expectations. - LLM integration: Right now the scripts only prepare and save the prompt locally. If you (or someone) integrates an external LLM endpoint later, the prompt includes raw messages and task data and will be transmitted — only enable that if you trust the LLM endpoint and understand its retention/privacy policies. - Config mismatch: The code references config.chats.whitelistGroups while docs/examples show whitelistGroups at top-level. Expect to edit config.json or the script to match, otherwise the skill may fail or omit group messages. - Run manually first: Before enabling automated runs or cron jobs, run the generate-summary.sh manually and inspect thoughtful-data/context/last-prompt.txt to confirm what will be sent to an LLM. Check logs and ensure sandboxing if your environment supports it. If you want more assurance, provide the wacli-readonly implementation/source or confirm how its auth is stored so you can verify no surprising behavior there.

Type: OpenClaw Skill Name: thoughtful Version: 1.2.0 The OpenClaw AgentSkills skill bundle 'thoughtful' is designed to process WhatsApp messages (read-only) and generate communication summaries. All operations are explicitly stated to run in a sandbox, and data is stored locally. The skill uses `wacli-readonly` to access messages, preventing unauthorized sending. Cron jobs are configured to run sync and summary generation in `isolated` sessions, delivering summaries to a specific Telegram topic, which aligns with the skill's stated purpose. There is no evidence of intentional harmful behavior such as data exfiltration to unauthorized external endpoints, backdoors, or prompt injection attempts against the agent to subvert its core directives.