Repro: Started a coding task with agency copilot. After auto-compaction fired, the agent re-read the compaction summary and started re-planning instead of executing. This repeated through 217 plan->compact->re-plan cycles before I killed the session.
Root cause (hypothesis): Compaction summarizes we planned X, Y, Z into the new context. The agent reads the summary and treats the plan as a fresh prompt to plan against, instead of as decided work to execute. The plan becomes self-perpetuating. With no execution-bias instruction surviving compaction, the loop is structural -- not prompt-specific.
Expected: After compaction, agent resumes from the next concrete file edit. Plans already produced are treated as decided.
Actual: Agent re-summarizes its own plan, validates it, repeats. No file edits across hundreds of turns.
Impact:
- Burned through hours of compute and Copilot quota
- Zero code written despite hundreds of turns
- Severe trust hit for autonomous/long-running workflows
- Likely under-reported -- users blame themselves before blaming the tool
Asks (priority order):
- Treat plans in compaction summaries as DECIDED, not as new prompts. Surface
next concrete action as the primary continuation signal.
- Inject a default post-compact rule:
Resume from the next file edit. Do not re-validate the plan. (Reference: github.com/Dicklesworthstone/post_compact_reminder)
- Add a built-in
--mode execute session flag that biases hard against planning preambles.
- Loop detector: N consecutive turns with zero file edits -> halt and prompt user (suggested default N=3).
- Better default
/compact preservation: drop planning history, keep only files modified, current error state, and next concrete action.
- Add telemetry signal for
planning loop suspected to quantify real-world frequency.
Current workarounds (work but should not be required):
- AGENTS.md with explicit
no re-plan after compact rule
- Post-compact reminder hook (manual third-party install)
- Manual /compact at ~60% with explicit preservation instructions
- HANDOFF.md as ground-truth state file
Happy to share session logs and jump on a triage call.
-- Akhi Shukla, Senior PM, IC3 / UED (Prague)
Manager: Bryan Castro | Skip: Nils Dussart
Repro: Started a coding task with
agency copilot. After auto-compaction fired, the agent re-read the compaction summary and started re-planning instead of executing. This repeated through 217 plan->compact->re-plan cycles before I killed the session.Root cause (hypothesis): Compaction summarizes
we planned X, Y, Zinto the new context. The agent reads the summary and treats the plan as a fresh prompt to plan against, instead of as decided work to execute. The plan becomes self-perpetuating. With no execution-bias instruction surviving compaction, the loop is structural -- not prompt-specific.Expected: After compaction, agent resumes from the next concrete file edit. Plans already produced are treated as decided.
Actual: Agent re-summarizes its own plan, validates it, repeats. No file edits across hundreds of turns.
Impact:
Asks (priority order):
next concrete actionas the primary continuation signal.Resume from the next file edit. Do not re-validate the plan.(Reference: github.com/Dicklesworthstone/post_compact_reminder)--mode executesession flag that biases hard against planning preambles./compactpreservation: drop planning history, keep only files modified, current error state, and next concrete action.planning loop suspectedto quantify real-world frequency.Current workarounds (work but should not be required):
no re-plan after compactruleHappy to share session logs and jump on a triage call.
-- Akhi Shukla, Senior PM, IC3 / UED (Prague)
Manager: Bryan Castro | Skip: Nils Dussart