CtrlK
BlogDocsLog inGet started
Tessl Logo

testland/nightvision-dast

Configures and runs NightVision white-box-assisted DAST: analyzes source code before attacking, traces every finding to its origin line, and drives coverage from OpenAPI / Postman / GraphQL specs rather than crawling. Supports Header, Cookie, TOTP, and recorded Interactive Login auth; exports findings as SARIF for GitHub Code Scanning, plus JSON, CSV, or PDF. Per-finding suppression via Alert Rules; CLI integration via the `nightvision` command. Use when source-traceable findings and spec-driven request coverage matter, not just authenticated black-box scanning (see zap-authenticated-scans for that).

75

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

Overview
Quality
Evals
Security
Files

Quality

Content

85%

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

A well-structured body with executable commands, an explicit scan-completion validation loop, and a clean one-level-deep reference split. The only weakness is moderate verbosity in the doc-quote framing and Overview exposition that could be tightened without losing signal.

Suggestions

Trim the Overview blockquote and the repeated 'Per [nv-docs][nv-docs] the platform supports:' preambles — keep the tables and commands, drop the narrative framing, to lift conciseness toward the lean anchor.

Collapse the Overview to a one-line positioning statement (white-box-assisted DAST with source-traceable findings) pointing to the docs link, since the description already covers this ground.

Reduce the prose intros above the Target types and Authentication tables to a single line each so the tables carry the information directly.

DimensionReasoningScore

Conciseness

Mostly efficient (tables, copy-paste commands, anti-patterns earn their place) but the Overview blockquote and repeated 'Per [nv-docs][nv-docs] the platform supports:' preambles add narrative tokens that could be trimmed, landing between the lean and the slightly-padded anchors.

2 / 3

Actionability

Provides fully executable, copy-paste-ready commands in the worked example (nightvision scan create with --spec/--auth/--auth-header, scan get --wait, scan results --output json/sarif) plus concrete install and target-type tables, matching the score-3 anchor.

3 / 3

Workflow Clarity

A clear 4-step 'How to use' sequence pairs with an explicit validation checkpoint and feedback loop ('the run must reach a completed state before you export. If --wait returns a failed or timed-out run ... fix ... and re-run'), so the destructive-scan workflow is not capped at 2.

3 / 3

Progressive Disclosure

SKILL.md is an overview that points to a single one-level-deep, well-signaled reference (references/ci-and-scan-operations.md, which exists on disk) for CI wiring, scope control, and output formats, matching the clear-overview-with-one-level-deep-references anchor.

3 / 3

Total

11

/

12

Passed

Description

100%

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 dense, third-person description that names concrete capabilities, surfaces natural trigger terms, gives an explicit Use-when clause, and explicitly disambiguates from a sibling skill. It is strong on all four dimensions with no first/second-person voice to penalize.

DimensionReasoningScore

Specificity

Lists multiple concrete actions ('analyzes source code before attacking, traces every finding to its origin line', 'drives coverage from OpenAPI / Postman / GraphQL specs', auth modes, 'exports findings as SARIF ... plus JSON, CSV, or PDF', 'Per-finding suppression via Alert Rules'), matching the score-3 anchor for several specific concrete actions.

3 / 3

Completeness

Explicitly answers both what (configures and runs NightVision DAST with traceable findings and spec-driven coverage) and when via a clear 'Use when source-traceable findings and spec-driven request coverage matter ...' clause, matching the score-3 anchor.

3 / 3

Trigger Term Quality

Covers natural domain terms a security practitioner would say — 'DAST', 'SARIF', 'GitHub Code Scanning', 'OpenAPI / Postman / GraphQL', 'black-box scanning', 'auth' — giving good trigger coverage rather than abstract jargon.

3 / 3

Distinctiveness Conflict Risk

Carves a clear niche (white-box-assisted DAST with source traceability) and explicitly redirects the overlapping case ('not just authenticated black-box scanning (see zap-authenticated-scans for that)'), making conflict with sister skills unlikely.

3 / 3

Total

12

/

12

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

Table of Contents