No. Keep durable preferences and facts in MEMORY.md and USER.md. Use Fulcra for state that changes often, backups you can restore, and anything other agents need to read. See How do I give OpenClaw or Hermes persistent memory?
Yes, if both connect to the same Fulcra account. Give each its own folder, for example /agents/hermes/ and /agents/openclaw/, and let each read the other's. See Shared memory for multiple agents.
Have the new agent read the backup from Fulcra at startup. Every upload is a new version, so you can pick an earlier one with file stat and file restore. See Back up and restore agent memory.
Each subagent can query Fulcra when it runs and pull only what its task needs, such as your calendar load for a scheduling subagent or your timezone for a research one. Neither sees the other's data. See Where should an agent store durable files and notes?
Start each run with uvx fulcra-api data-updates "1 hour" (or the window since its last run). It lists what changed, so Hermes reads only that. See How an agent finds out what changed since its last run.
Yes. The scheduled job writes its results to a Fulcra file such as /agents/nightly/state.md, your chat writes requests to its own file, and each runs data-updates at the start to see which files changed. Read the full answer.
Yes. Hermes has no endpoint for other agents to call, so Fulcra's connect-our-agents skill sets up a channel you both opt into, and nothing else in either account is shared. Read the full answer.
.png){kind=logo}
