| name | ecc-cmd-loop-start |
| description | Command skill for 'loop-start'. Use when users request bounded iterative execution cycles; initialization and launch flows. |
ecc-cmd-loop-start
Initialize a bounded execution loop with explicit controls.
Required Inputs
- Objective
- Max iterations
- Stop conditions
- Verification mode (
quick|full)
Workflow
- Define loop contract (goal, bounds, stop rules).
- Create iteration template:
- plan step
- execute step
- verify step
- checkpoint summary
- Start at iteration 1.
- Stop immediately on blocker/critical failure.
- End with final checkpoint.
Output
## Loop Started
- Objective: ...
- Max iterations: N
- Stop conditions: ...
- Verification: quick|full
- Status: running
Rules
- Never run unbounded loops.
- Always report current iteration and latest verdict.
ECC Parity Notes
- This skill is adapted from ECC intent and translated to OpenClaw-native tool workflows.
- Prefer reproducible, tool-backed execution over harness-specific slash semantics.
- Keep outputs evidence-based and verification-first.
OpenClaw Constraints
- If upstream flow assumes unavailable hooks/tools, use documented fallback scripts in
openclaw-ecc/scripts/.
- Mark non-applicable checks as
SKIP rather than hard-failing when environment support is absent.
