resume

name	resume
description	Resume an interrupted task from .dynos/ state. Use after session restart or context compression.

dynos-work: Resume

Resume a dynos-work task that was interrupted.

Ruthlessness Standard

• Do not guess the next command from vibes; derive it from the actual task stage.
• Missing or corrupt artifacts are not "probably fine". Call them out.
• If multiple active tasks exist, do not silently pick the wrong one.

What you do

1. List all tasks in .dynos/ that are not DONE or FAILED (read each manifest.json). If available in this repo, use "${CODEX_PLUGIN_ROOT:-${CLAUDE_PLUGIN_ROOT:-}}/bin/dynos" ctl active-task.
2. If one active task: resume it automatically
3. If multiple active tasks: show list and ask which to resume
4. Read manifest.json to determine current stage
5. Tell the user which command to run next based on the stage:

Stage	Run this
SPEC_NORMALIZATION, SPEC_REVIEW	/dynos-work:start
PLANNING, PLAN_REVIEW, PLAN_AUDIT	/dynos-work:plan
PRE_EXECUTION_SNAPSHOT, EXECUTION, TEST_EXECUTION	/dynos-work:execute
CHECKPOINT_AUDIT, REPAIR_PLANNING, REPAIR_EXECUTION, FINAL_AUDIT	/dynos-work:audit

Output on resume

dynos-work: Resuming task-20260327-001
Title: [task title]
Current stage: CHECKPOINT_AUDIT

Run: /dynos-work:audit

Edge cases

• If manifest.json is corrupt or unreadable: report the error and suggest starting fresh with /dynos-work:start
• If execution-graph has segments with no evidence files: /dynos-work:execute will re-run those segments
• If audit-reports exist from before interruption: /dynos-work:audit will read them and determine if re-audit is needed

When to use

• After a session restart during a long task
• After Claude Code context was compressed mid-task
• After a network interruption
• After manually investigating the task state with /dynos-work:status