OpenClaw Flomo Skill

Read and write flomo memos on macOS. Use when user asks to fetch recent flomo notes, search memo text in local flomo cache, or create/write a new memo to flo...

This skill appears to do what it says: to work it reads your local flomo desktop data (config.json and IndexedDB) to obtain the access_token and to discover incoming webhook paths, and it calls flomo APIs/webhooks. That access is necessary for functionality but is sensitive: do not share your flomo config.json. Before installing, verify you trust the skill source and review scripts/flomo_tool.py yourself. Note the registry metadata does not declare required binaries (the script invokes curl and strings) or the optional env vars documented in README; ensure you are comfortable giving the script read access to ~/Library/Containers/com.flomoapp.m/... and network access to flomoapp.com. If you prefer tighter control, set FLOMO_ACCESS_TOKEN or FLOMO_CONFIG_PATH to a limited/throwaway token or run the script in a confined environment and inspect its output (it masks webhook URLs when printing).

Type: OpenClaw Skill Name: openclaw-flomo-skill Version: 1.0.0 The OpenClaw Flomo skill is designed to interact with the local Flomo macOS application, reading its configuration and using its API for memo management. The `scripts/flomo_tool.py` accesses local Flomo application files (`config.json`, `IndexedDB`, `renderer.log`) to retrieve necessary tokens and state, which is transparently documented in `README.md` as part of its security model. Network calls are made to `flomoapp.com` using `curl` via `subprocess.run` with arguments passed as lists, mitigating shell injection risks. User-provided content for memo writing is properly JSON-encoded or URL-encoded. The `SKILL.md` instructions guide the AI agent appropriately, including a positive security practice of masking webhook URLs. There is no evidence of intentional data exfiltration to unauthorized endpoints, malicious execution, persistence mechanisms, or obfuscation. The capability to specify a webhook URL could be misused by a malicious user, but the skill itself does not exhibit malicious intent.