CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

reverse-skill-router

Routes reverse engineering, exploitation, penetration testing, malware, mobile, firmware, browser automation, documentation, and security tasks to the appropriate specialist skill. Use when a task spans modules or the correct reverse-skill entrypoint is unclear.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Failed to scan

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./skills/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

77%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.

A well-engineered orchestrator document: tightly written, with a gated and ordered execution contract, concrete copy-paste commands, failure branches, and a completion checklist. The main defects are instructional sections that assume a bundle larger than what actually ships (missing routing config, tool-index, ops and journal files) and a few trimmable meta sections.

Suggestions

Ship or generate the files the contract hard-depends on (config/routing.json, tool-index.md) in the bundle, or state explicitly that case-init/bootstrap create them on first run, so the first NOW step is executable as bundled.

Remove the RFC 2119 semantics section and condense the "目录是动态扩充的" / "自动进化" housekeeping into one short pointer, cutting tokens that do not instruct the current task.

Verify every referenced path (ops/*.md, RULES.md, CONTRIBUTING.md, field-journal/) exists in the published bundle, or mark them as created-on-demand so navigation does not dead-end.

DimensionReasoningScore

Conciseness

The body is dense and telegraphic — the routing contract, module table, and bootstrap commands earn their tokens — but a few sections could be trimmed: the RFC 2119 MUST/SHOULD/MAY semantics Claude already knows, and meta-commentary sections like "目录是动态扩充的" and "自动进化" that describe repository housekeeping rather than instructing the task.

4 / 5

Actionability

Concrete, copy-paste-ready commands are given for the core path (master-route.ps1/.sh, case-init, bootstrap-reverse with full command lines and placeholders) plus a concrete next-step menu example. The gap: the very first NOW step depends on config/routing.json and downstream steps on tool-index.md, neither of which exists in the provided bundle, so the primary instructions are not fully executable as bundled.

4 / 5

Workflow Clarity

The CRITICAL contract gives a strictly ordered NOW/NOW/ACT/NEXT sequence with a hard authorization gate ("auth 未 granted 禁止对目标 ACT"), explicit error-recovery branches (routing miss → research and propose a new skill, never force-fit; missing tool → bootstrap), a read-order table for precedent files, and a terminal self-check checklist — matching the anchor for clear sequence with explicit validation, feedback loops, and checklists.

5 / 5

Progressive Disclosure

References are clearly signaled with their purpose and stay one level deep (scripts, references/*.md, ops/*, field-journal/*), and the 40-row module table is appropriately inline for a router index. However, scored against the actual bundle, 7 of 16 referenced paths are absent (config/routing.json, routing.md, tool-index.md, ops/IDENTITY.md, RULES.md, CONTRIBUTING.md, field-journal/), so a reader following the contract immediately hits broken navigation — more than a minor organization gap.

3 / 5

Total

16

/

20

Passed

Description

71%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 solid router description: concrete action, good natural-domain keyword coverage, and an explicit Use-when clause. Its weaknesses are an internal-jargon trigger condition, missing common synonyms (CTF, pentest, decompile), and a overly broad trailing scope ("documentation, and security tasks") that raises conflict risk with the very skills it routes to.

Suggestions

Rewrite the trigger clause in user-facing language, e.g. "Use when the user mentions reverse engineering, CTF, pentesting, malware analysis, APK/binary analysis, or when a task spans multiple of these areas and the right entrypoint is unclear."

Trim the overly broad trailing terms ("documentation, and security tasks") to the specific domains the router actually dispatches, reducing conflict with general documentation/security skills.

Add high-frequency synonyms users actually say — CTF, pentest, decompile, APK, binary analysis, exploit development — to improve trigger term coverage.

DimensionReasoningScore

Specificity

"Routes reverse engineering, exploitation, penetration testing, malware, mobile, firmware, browser automation, documentation, and security tasks to the appropriate specialist skill" states one concrete action (routing/dispatch) with a broad enumeration of target domains. Minor gaps remain (forensics, cloud/K8s, CTF, OSINT covered by the skill's body but unnamed here), so it falls at 'several specific actions with minor gaps' rather than comprehensive.

4 / 5

Completeness

Both parts are present: the what ("Routes ... tasks to the appropriate specialist skill") and an explicit when ("Use when a task spans modules or the correct reverse-skill entrypoint is unclear"). The when-clause is explicit but narrow and system-internal — it only covers the ambiguous-routing case rather than concrete user triggers, so it does not reach the fully explicit level.

4 / 5

Trigger Term Quality

Natural keywords like "reverse engineering", "penetration testing", "malware", "firmware", and "browser automation" match what users would actually say, but common variations are missing (CTF, pentest, decompile, APK, binary analysis), and "the correct reverse-skill entrypoint is unclear" is internal jargon rather than a user-facing phrase.

4 / 5

Distinctiveness Conflict Risk

The trailing "documentation, and security tasks" is extremely broad and could collide with general documentation or security skills, and as a master router it deliberately fronts the same trigger terms as every specialist skill it dispatches to. It is somewhat specific (clearly a reverse-engineering/security dispatcher) but overlap risk with the family of skills it routes to is real.

3 / 5

Total

15

/

20

Passed

Validation

100%

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/SKILL.md

Reviewed

Table of Contents

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.