CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

draft-security-advisory

Draft a GitHub security advisory (GHSA) for nrwl/nx from a vulnerability report. Investigates the affected code, determines affected packages and version ranges, proposes a CVSS score and CWEs, and writes a draft advisory following the layout of previously published nrwl/nx advisories. Output is a local draft only — a human reviews and publishes. Use when a security vulnerability report needs to become an advisory, or when the user says "draft an advisory", "write a GHSA", or "security advisory for <report>".

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 strong, highly actionable instruction-only skill: concrete git/gh/npm commands, an explicit five-phase workflow with stop conditions, and a complete output template. The only weaknesses are minor duplicated emphasis of the safety rules and a monolithic layout template that could be split out.

Suggestions

State the never-publish/human-review boundary once (e.g., in Hard rules only) and let the intro reference it in a few words, removing the duplicated sentences about publishing and verification.

Move the ~50-line advisory body layout template into a references/ file (e.g. references/advisory-template.md) and point to it from Phase 5, keeping SKILL.md as a lean overview.

Merge the 'grounded in the actual code, not just the reporter's claims' guidance in the Process preamble with the equivalent Hard rules bullet to eliminate the remaining redundancy.

DimensionReasoningScore

Conciseness

The body is dense and assumes Claude's competence (no explanations of what CVSS or a GHSA is), but the never-publish/human-review rule is stated in both the intro and Hard rules, and the verify-don't-copy point appears in both the Process preamble and Hard rules — minor duplication that could be trimmed, so below the 'every token earns its place' anchor.

4 / 5

Actionability

Fully executable guidance throughout: copy-paste commands with placeholders (`git log -S '<code fragment>'`, `git tag --contains <sha> | sort -V | head`, `npm view <pkg> versions --json`, `gh api /repos/nrwl/nx/security-advisories/<GHSA-ID>`), an exact output path, and a complete advisory layout template that covers the common case end to end.

5 / 5

Workflow Clarity

Five phases explicitly sequenced with a 'do not skip the investigation phases' directive and explicit validation checkpoints: stop-and-report if the flaw cannot be confirmed, npm verification that patched versions are published, and a closing recap for reviewer sanity-check. This matches the clear-sequence-with-explicit-validation anchor; the destructive/batch cap does not apply since output is a local draft only.

5 / 5

Progressive Disclosure

A single well-organized ~145-line file with clear section headers, no buried references, and no bundle files to navigate — good structure with everything appropriately placed. It falls short of the level-5 anchor only because it exceeds the under-50-line simple-skill case, and the ~50-line advisory layout template is content that could live in a references/ file as the skill grows.

4 / 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.

An exemplary description: concrete third-person action verbs, an explicit output boundary, and verbatim natural trigger phrases. It clearly separates itself from neighboring skills by scoping to nrwl/nx advisory drafting.

DimensionReasoningScore

Specificity

Lists multiple concrete actions covering the full workflow — "Investigates the affected code", "determines affected packages and version ranges", "proposes a CVSS score and CWEs", "writes a draft advisory following the layout of previously published nrwl/nx advisories" — with no coverage gaps; level 4 would require minor gaps.

5 / 5

Completeness

Explicitly answers both questions: 'what' via five concrete actions plus the output boundary ("Output is a local draft only — a human reviews and publishes"), and 'when' via "Use when a security vulnerability report needs to become an advisory, or when the user says..." with concrete trigger phrases.

5 / 5

Trigger Term Quality

Quotes the exact natural phrases a user would say — "draft an advisory", "write a GHSA", "security advisory for <report>" — plus synonyms (GHSA, advisory, vulnerability report); this is the comprehensive-coverage anchor rather than the 'a few natural terms missing' level 4.

5 / 5

Distinctiveness Conflict Risk

Pinned to a single repository ("for nrwl/nx") and a narrow artifact type (a local GHSA draft), giving it a clear niche with distinct triggers and minimal conflict risk with any other skill.

5 / 5

Total

20

/

20

Passed

Validation

87%

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

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository

nrwl/nx

Path

.claude/skills/draft-security-advisory/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.