radare2

Use this skill whenever the user wants to analyze binaries with radare2/r2 from the command line, including reverse engineering, disassembly, function analysis, strings/import inspection, patching, binary diffing, hex inspection, or r2 scripting. Also use it when the user mentions PE/ELF/Mach-O/DEX/WASM files together with CLI analysis, `rabin2`, `rasm2`, `radiff2`, `r2pipe`, or asks for radare2 command help on Windows/Linux/macOS.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%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.

The body is highly actionable with well-sequenced workflows, verification steps, and a completion checklist, but it is not maximally concise: it duplicates the cheatsheet's command reference inline and carries process/meta overhead (external precedent references, routing context, a redundant bootstrap table). Progressive disclosure is decent but undermined by that duplication.

Suggestions

Move the per-tool command reference ('常用子工具' sections for rabin2/rasm2/radiff2/rahash2/rax2) into references/cheatsheet.md and keep SKILL.md as a lean overview that links out, reducing inline duplication.

Trim process meta-overhead that does not help execute the task — the 'ACTION REQUIRED' block referencing ../field-journal/precedent-reverse.md and ../tool-index.md, the '路由上下文' block, and the bootstrap table that restates behavior already encoded in recon.ps1.

Tighten the patch workflow into an explicit validate→fix→retry feedback loop (e.g., 'after wa/wx, run pdf to confirm; if wrong, re-edit and re-verify before wq') to make the destructive-operation checkpoints unambiguous.

Dimension	Reasoning	Score
Conciseness	Mostly command-focused and efficient, but carries organizational meta-overhead ('ACTION REQUIRED' referencing external precedent files, routing context, a bootstrap table duplicating what recon.ps1 already does, task self-check) and duplicates cheatsheet command listings inline. Not a 1 because it avoids explaining concepts Claude already knows; not a 3 because these tokens do not all earn their place.	2 / 3
Actionability	Provides fully executable, copy-paste-ready commands throughout ('rabin2 -I', 'r2 sample.exe', 'aaa', 'pdf', 'wa nop', 'wq') plus real script invocations with parameters. Not a 2 because the guidance is concrete and complete rather than pseudocode or abstract.	3 / 3
Workflow Clarity	Six numbered workflows plus a numbered recommended analysis order, a completion self-check checklist, explicit patch verification ('修改后再次反汇编验证'), and bootstrap failure handling. Not a 2 because validation/verification steps for the destructive patch operation are present rather than missing.	3 / 3
Progressive Disclosure	Bundle files (references/cheatsheet.md, scripts/recon.ps1, scripts/recon.sh) are one-level-deep and signaled via the '内置资源' and '参考资料' sections, but the body duplicates the cheatsheet command reference inline rather than staying a lean overview. Not a 1 because references are real and clearly signaled; not a 3 because content that should be separate is duplicated inline.	2 / 3
	Total	10 / 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.

The description is strong across all dimensions: concrete actions, natural trigger terms with named subtools, explicit what-and-when guidance, and a clearly distinct niche. It uses appropriate third-person voice with no over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'reverse engineering, disassembly, function analysis, strings/import inspection, patching, binary diffing, hex inspection, or r2 scripting' — matching the anchor for several specific actions. Not a 2 because the action list is comprehensive rather than partial.	3 / 3
Completeness	Explicitly answers both what (binary analysis actions) and when ('Use this skill whenever the user wants to analyze binaries... Also use it when the user mentions...'), with third-person voice. Not a 2 because the 'when' is an explicit trigger clause, not merely implied.	3 / 3
Trigger Term Quality	Covers natural terms users say ('analyze binaries with radare2/r2 from the command line') plus tool names they mention ('rabin2', 'rasm2', 'radiff2', 'r2pipe') and file types ('PE/ELF/Mach-O/DEX/WASM'). Not a 2 because it spans common variations rather than a single keyword.	3 / 3
Distinctiveness Conflict Risk	Tool-specific triggers (rabin2/rasm2/radiff2/r2pipe, radare2/r2 CLI) carve a clear niche unlikely to collide with other skills. Not a 2 because the named subtools make it highly distinguishable.	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: zhaoxuya520/reverse-skill
Path: skills/radare2/SKILL.md
Commit: 6aa1362

Reviewed: 7 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.