Agents In UseAn illustrative video entry showing how an agent workflow could be reviewed, with a deliberate human checkpoint.
OpenClawThis example describes a possible use case; no real-world result has been reported.
A person sets the scope and access permissions, reviews the final output, and approves any external publication or consequential action.
A person approved each step before the agent continued.
Illustrative workflow: after receiving a goal and bounded access, the agent plans the work, executes multiple steps, checks its output, and prepares a deliverable without prompts at each step.
EXAMPLE ONLY. The video URL and author are placeholders, not real sources. No recording or deployment is claimed.
Credit: Agents In Use editorial example
OpenClaw · Approved workspace tools
Code · Testing
Editorial completeness: 85/100. Measures documented fields, not agent performance or verified outcomes.
@example · X
til.simonwillison.netWillison ran o4-mini-deep-research to find surviving orchestrions and return structured location records. The model used web search and code execution; he explicitly considered the resulting list incomplete and unverified.
Added
OpenAI deep research
futurehouse.orgFutureHouse used Robin to connect literature research, experiment planning and analysis for dry age-related macular degeneration. Human researchers ran the lab work; the system proposed follow-up candidates from the results.
Added
Robinfuturehouse.orgFutureHouse ran WikiCrow to draft scientific summaries of human genes using full-text papers. Its PaperQA agent searched, gathered evidence and returned to retrieval when terminology or evidence required it.
Added
WikiCrow