openclaw-debugger

Trigger when an OpenClaw agent is broken, silent, crashing, stuck, not responding, returning empty output, or the user says "my agent is down", "agent not working", "/openclaw-debugger". Walks through the standard OpenClaw 2026.4 diagnosis checklist and prints a report.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

65%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced diagnostic checklist with concrete commands and thresholds, weakened mainly by hardcoded time-sensitive version pins and a missing destructive-step re-validation loop. A dangling TROUBLESHOOTING.md reference also keeps progressive disclosure from scoring top marks.

Suggestions

Move the OpenClaw 2026.4.11 compatibility note and pinned model versions (Opus 4.6, Sonnet 4.6, GLM-5.1, Minimax M2.7) into an explicit version/compatibility or 'supported versions' section so time-sensitive details don't dilute the core checklist.

Add a validate-after-fix feedback loop for the destructive session-file step (e.g. re-run `openclaw agent --status` or a json.tool check after the backup-and-delete) so the workflow clears the destructive-operation validation bar.

Either create the referenced TROUBLESHOOTING.md in the bundle or remove the dangling reference so progressive disclosure navigation resolves to a real file.

Dimension	Reasoning	Score
Conciseness	The body is largely lean and terse with no concept-explaining fluff, but hardcodes time-sensitive version pins ("Compatible with OpenClaw 2026.4.11" and model versions like Opus 4.6, GLM-5.1) outside any deprecated/old-patterns section, which the guidelines penalize. It is mostly efficient but could be tightened by isolating version specifics.	2 / 3
Actionability	Every step ships concrete, copy-paste-ready commands (e.g. `openclaw agent --status`, `lsof -i :18789`, `python3 -m json.tool < sessions.json`, `curl localhost:11434/api/tags`) with precise thresholds (5 min, 60s, 5 MB) and matching fixes, fully matching the executable-guidance anchor.	3 / 3
Workflow Clarity	The six steps are clearly numbered and gated by an explicit "Stop at the first failing step" checkpoint, but the destructive session-file deletion (mv ...{,.bak}) lacks a re-validation feedback loop, and the guidelines cap workflow clarity at 2 when validation/verification steps are missing for destructive operations.	2 / 3
Progressive Disclosure	Sections are reasonably well organized, but the body points to a deeper reference ("see TROUBLESHOOTING.md in this repo") that does not exist in the bundle, leaving a dangling one-level reference with no actual backing file, so it is not a clean 3.	2 / 3
	Total	9 / 12 Passed

Description

100%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, well-targeted description that explicitly covers what the skill does and when to invoke it with natural trigger terms and a clearly distinct niche. It is concise, concrete, and uses appropriate third-person voice.

Dimension	Reasoning	Score
Specificity	Lists many concrete failure states ("broken, silent, crashing, stuck, not responding, returning empty output") and a concrete outcome ("Walks through the standard OpenClaw 2026.4 diagnosis checklist and prints a report"), matching the anchor that lists multiple specific concrete actions.	3 / 3
Completeness	Clearly answers both what it does (diagnosis checklist and report) and when to trigger it ("Trigger when an OpenClaw agent is broken...") with explicit trigger guidance and third-person voice, so it is not capped at 2.	3 / 3
Trigger Term Quality	Includes natural phrasings users would actually say ("my agent is down", "agent not working") plus the slash-command trigger "/openclaw-debugger", giving good coverage of natural terms rather than just technical jargon.	3 / 3
Distinctiveness Conflict Risk	The OpenClaw 2026.4 agent-diagnosis niche with product-specific trigger terms is clearly distinct and unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: mergisi/awesome-openclaw-agents
Path: skills/claude/openclaw-debugger/SKILL.md
Commit: 05820c5

Reviewed: 8 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.