debugging

MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'playwright', 'flaky test', 'fails intermittently', 'passes in isolation', 'only fails in CI'.

Quality

92%

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

85%Weight 40%Scale 1-5

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

An exemplary progressive-disclosure index: a lean, well-signaled router to one-level-deep references with a clearly sequenced, validation-gated phase loop. Slightly held back by repeated gate-rule emphasis and by delegating most executable detail to the references.

Suggestions

Consolidate the 'read the reference first' gate rule to a single authoritative statement; it is currently repeated in the intro, the callout block, safety invariant #8, and the closing checklist.

Inline one tiny canonical example per runtime row (e.g. a one-line attach command) so the index is directly executable for the most common case without requiring a reference round-trip.

Dimension	Reasoning	Score
Conciseness	The body is a lean index that mostly assumes Claude's competence, but the gate-rule ("read the reference first") is restated in the intro, the warning callout, safety invariant #8, and the closing "What to Do Right Now" section, adding repetition that could be trimmed. Not a 5 because of this repeated emphasis; not a 3 because the bulk is efficient and non-padded.	4 / 5
Actionability	As an index it delegates executable how-to to references, but it still provides concrete commands ("file ./target", "du -h ./target", "strings -n 12 ./target \| rg -iE '...'"), a native-vs-bundled discriminator, and a numbered "What to Do Right Now" sequence. Not a 5 because the core executable guidance lives in the references rather than inline; not a 3 because the inline guidance is concrete and actionable.	4 / 5
Workflow Clarity	The phase loop (0–10) is explicitly sequenced with one reference per phase, backed by safety invariants, the red→green test gate, the Oracle Triple feedback loop after two failed rounds, and the four evidence gates in Final verification — a clear sequence with explicit validation steps and error-recovery feedback loops.	5 / 5
Progressive Disclosure	SKILL.md is a well-organized overview/index and all 19 referenced paths (runtimes, tools, methodology) resolve to real files exactly one level deep, with tables that map each scenario to its reference for easy navigation.	5 / 5
	Total	18 / 20 Passed

Description

100%Weight 40%Scale 1-5

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, specific, well-triggered description that names concrete actions, exhaustively lists natural trigger phrases and debugger tools, and clearly separates what the skill does from when to invoke it. Uses third-person voice throughout with no fluff.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — "form ≥3 hypotheses", "investigate in parallel", "spawn Oracles from orthogonal angles", "confirm root cause", "lock with a failing test", "fix minimally", "QA by actually USING the system", "scrub artifacts" — giving comprehensive coverage of the skill's behavior.	5 / 5
Completeness	It explicitly answers both "what" (a hypothesis-driven loop with named steps) and "when" ("MUST USE for any real runtime debugging across ANY language or binary" plus an explicit "Triggers:" list with concrete phrases).	5 / 5
Trigger Term Quality	Extensive natural trigger phrases users would actually say ("debug this", "why is X not working", "hanging", "flaky test", "silent failure", "HTTP 200 but empty") plus debugger tool names (gdb, lldb, pdb, dlv, pwndbg), giving comprehensive coverage including synonyms and tool invocations.	5 / 5
Distinctiveness Conflict Risk	It carves a clear niche — real runtime debugging and reverse engineering across any language/binary — with a distinctive hypothesis-driven methodology and specific tool triggers, minimizing overlap with general coding skills.	5 / 5
	Total	20 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 24 deeper-than-1-level	Warning
referenced_paths_exist	Referenced path issues: 49 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: code-yeongyu/lazycodex
Path: plugins/omo/skills/debugging/SKILL.md
Commit: 10f9558

Reviewed: 10 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.