Unverified50% confidenceFactExact time
Resumption in the governed runtime is purely templated and model-free: it completes the remainder of the flow without calling the reasoner again; if any of the remaining steps calls a model, none of them run, and instead it waits for the user's next message.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe resumption gating of the governed runtime is only as strict as the flow policy, and only takes effect on the approval path; flows that declare no resume policy resume for anyone. When a paused run is instead picked up from the user's next chat message, the runtime reads the session's role snapshot rather than re-reading assignments, and on denial detaches the run from the conversation, keeping it in the paused state without failing it.77% similarUnverifiedWhen resuming approval, what should be persisted is the pending call rather than the agent's memory, and on resume the saved call should be executed directly rather than having the model reason again.74% similarUnverifiedPlan mode is no longer lost when resuming with --continue or --resume <session-id>, and restored sub-agents and teammates no longer lose their prior reasoning processes and prompt cache.72% similarUnverifiedThe model itself is stateless; state management can be achieved through structured To-Do List progress tracking and Git checkpoints to enable long tasks to be recoverable and resumable.71% similarUnverifiedRule-based fallback means that when an external dependency fails, it falls back to a pre-written set of templates or a mapping table to generate output; its readability is inferior to explanations generated by a large model, but it is stable, controllable, zero-latency, and zero-cost.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1048459API
curl https://kongchang.com/api/v1/knowledge/claims/1048459MCP
get_claim(id=1048459)