Concept

Human re-entry

You might say: escalation point, handoff boundary

The point where a person comes back after a run ends or a gate blocks progress. The person's change or decision becomes a new event that starts another run.

What it does

Keeps automated work bounded while preserving a clear path for judgement, clarification, and recovery.

Ask

What exact evidence or failure brings a person back, and what event starts after their change?

What to look for

1. A stopped run with a human-review or needs-clarification label
2. A comment, approval, or ticket edit that starts a new run
3. A handoff record naming the decision still needed

Pattern

Stop the line

Connections

• See also: Try cap
• See also: Trigger

Sources

1. Anthropic, Building effective AI agents, AgentsAgents can then pause for human feedback at checkpoints or when encountering blockers.
2. Betsy Beyer et al., Site Reliability Engineering, Chapter 1, IntroductionMonitoring should never require a human to interpret any part of the alerting domain. Instead, software should do the interpreting, and humans should be notified only when they need to take action.
3. GitHub, Best practices for using GitHub Copilot to work on tasks, Using comments to iterate on a pull requestAfter a user with write access mentions `@copilot` in a comment, Copilot will start to make any required changes, and will update the pull request when it's done.

Use it on your project

teardown-apply: hold "Human re-entry" (human-re-entry) from the teardown at https://teardownatlas.com/software-factory/ against my project

Paste it into a coding agent that has the teardown skills.

An AI agent wrote this map with the teardown skill. It can contain errors: check the sources before you rely on a part. About and legal