Conciseness

Every sentence carries operational information — the timeout guidance ("normally 11 minutes"), the availability check ("command -v openclaw"), the payload-parsing caveat, and safety constraints — with no padding and no explanation of concepts Claude already knows. This matches the lean, every-token-earns-its-place anchor.

Actionability

The core guidance is concrete and executable: an exact command ("openclaw agent --local --agent main --message \"<prompt>\" --json --timeout 600") and a specific availability check ("command -v openclaw"). It falls just short of the top anchor because the JSON payload's success/failure structure is never shown, leaving "accept the response only when the JSON payload reports success" without a literal field-level example.

Workflow Clarity

The three-step sequence (set working directory and timeout, check availability, run once) includes explicit validation checkpoints — parse the JSON payload "even when the process exits zero" and "Report payload errors and timeouts as failures" — plus defined failure paths (stop and ask the user on missing binary or missing onboarding/credentials). This matches the explicit-validation anchor rather than the minor-gaps anchor at 4.

Progressive Disclosure

The body is under 50 lines, requires no external references (the bundle has no references/, scripts/, or assets/ directories), and is organized into two clear sections ("## Run", "## Authentication And Permissions"). Per the simple-skill guideline, this earns the top score with just well-organized sections.

Specificity

The description names the domain ("the local OpenClaw agent") and one concrete action ("Run ... non-interactively"), matching the anchor for 1-2 concrete actions without comprehensive coverage. It does not list several distinct specific actions that would merit a 4, but it is more concrete than the generic domain-only anchor at 2.

Completeness

It explicitly answers both parts: the what is "Run the local OpenClaw agent non-interactively" and the when is the concrete trigger clause "Use when the user asks to delegate a bounded task to OpenClaw". The when is already explicit and specific, so the 4 anchor ('when' could be more explicit or specific) does not fit.

Trigger Term Quality

Natural terms a user would say are present — "OpenClaw", "delegate a bounded task", "agent" — giving good keyword coverage. A few natural phrasings are missing (e.g. "ask OpenClaw", "hand off to OpenClaw"), so it fits the good-coverage-with-minor-omissions anchor rather than the comprehensive-synonyms anchor at 5.

Distinctiveness Conflict Risk

The description is anchored to a uniquely named tool ("OpenClaw") with a narrow, distinct trigger (delegating bounded tasks to that tool), giving a clear niche with minimal conflict risk. It goes beyond the 'mostly distinct' 4 anchor.