Fähigkeiten feeling-search
📦

feeling-search

v1.2.0 Inhaltsrevision r5 Mittleres Risiko ⚙️ Externe Befehle🌐 Netzwerkzugriff📁 Dateisystemzugriff🔑 Umgebungsvariablen

Find Visual Styles from Mood Descriptions

Everyday mood descriptions often lack precise art-direction vocabulary. This skill matches descriptions to ranked visual styles, definitions, and reference vocabulary through local matching and optional online lookups.

Unterstützt: Claude Codex Code(CC)
Maintain this Skill? Claim it 📊 74 Angemessen

Mit meinem Agent installieren

Kopieren Sie diese Anfrage in Ihren Agent. Sie enthält die maßgebliche Skill-Seite und das Manifest.

Agent-Anfrage
Review the Skillstore skill "feeling-search" from https://skillstore.io/skills/xixi048-feeling-search.md and its manifest at https://skillstore.io/api/skills/xixi048-feeling-search/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Ihr Agent sollte weiterhin seinen Plan anzeigen und alle von der Sicherheitsrichtlinie verlangten Bestätigungen anfordern.

Agent-lesbare Ressourcen

Verwenden Sie diese Links, wenn ein KI-Agent, Crawler oder Skript sauberen Kontext benötigt, statt die vollständige Seite zu lesen.

Testen

„feeling-search“ wird verwendet. Describe styles for the documented Chinese example about walking alone past a convenience store in rain under cold white lighting.

Erwartetes Ergebnis:

  • Documented local example: the matched terms include light and white.
  • Documented candidate styles include the Delft School, Ashen Winter, and De Stijl.
  • These are lexical candidates, not a definitive interpretation of the scene.

„feeling-search“ wird verwendet. Analyze a mood offline before a dictionary cache has been downloaded.

Erwartetes Ergebnis:

No local dictionary cache is available. Download the dictionary while online, then retry offline. No network request was made.

Sicherheitsaudit

Mittleres Risiko
v5 • 29.10.2026 Versionsbericht öffnen

Of 115 static findings, four are confirmed: unpinned package execution, insecure token-file creation, and credential forwarding in two HTTP helpers. The remaining alerts reflect documented client behavior, Markdown formatting, or readable multilingual text. A semantic finding covers quota tokens in URLs; no evidence found of malicious intent or prompt injection.

3
Gescannte Dateien
1,136
Analysierte Zeilen
4
Prüfelemente
0
Falschmeldungen ignoriert

Bestätigte Sicherheitsbedenken (1)

Mittel
Bearer Token Exposure Through Quota Query URLs
quota_url places the quota bearer token in a URL query parameter. Balance and token-validation requests can expose this reusable credential to access logs and monitoring.
The client builds credential-bearing URLs, and the contract accepts the same token in an authentication header. Actual server logging was not verified.
Elemente der Fähigkeitsprüfung (4)

Dies sind echte lokale Fähigkeiten, die für diese Fähigkeit erwartet werden können; daher müssen sie überprüft werden, werden jedoch nicht als bestätigtes bösartiges Verhalten gezählt.

Mittel
Python HTTP libraries
with urllib.request.urlopen(req, timeout=TIMEOUT) as resp:
urlopen follows redirects with custom headers added by http_get, including X-FS-Quota from style and images calls. Cross-origin redirects can disclose credentials.
Mittel
Python HTTP libraries
with urllib.request.urlopen(req, timeout=TIMEOUT) as resp:
The POST helper uses default redirect handling after adding caller-supplied headers. A redirect from cloud analysis can forward X-FS-Quota to another origin.
Mittel
Python file write/append
with open(tmp, "w", encoding="utf-8") as fh:
save_quota writes a bearer token without restrictive permissions to a predictable temporary path. With a permissive umask and traversable parent directories, other local users can read the saved credential.
Mittel
Ruby/shell backtick execution
```bash
The block instructs npx -y feeling-search-mcp@latest --self-test, downloading and executing an unpinned external package. Automatic confirmation creates a supply-chain execution risk.
Geprüft von: codex Audit-Verlauf anzeigen →
Diesen Bericht teilen & zitieren

Teile den versionierten Bewertungsbericht, das neutrale Badge, die Einbettungskarte und Zitate. Skillstore berichtet Nachweise, ohne zu entscheiden, ob dieser Skill sicher ist.

Versionsbericht öffnen
Sicherheitsbewertung

Berichtslink kopieren

https://skillstore.io/skills/xixi048-feeling-search/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown-Badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/xixi048-feeling-search/security.svg)](https://skillstore.io/skills/xixi048-feeling-search?utm_source=security_passport_badge)

HTML-Badge

<a href="https://skillstore.io/skills/xixi048-feeling-search?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/xixi048-feeling-search/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Einbettungskarte

<iframe src="https://skillstore.io/embed/skills/xixi048-feeling-search.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Wissenschaftliche Zitate (APA · BibTeX · CFF)

APA-Zitat

xixi048. (2026). feeling-search security audit report (audit version 5) [Author version 1.2.0]. Skillstore. https://skillstore.io/skills/xixi048-feeling-search/audits/5

BibTeX-Zitat

@techreport{xixi048-xixi048-feeling-search-2026, author = {xixi048}, title = {feeling-search security audit report (audit version 5)}, institution = {Skillstore}, year = {2026}, number = {5}, url = {https://skillstore.io/skills/xixi048-feeling-search/audits/5}, note = {Author version 1.2.0} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "feeling-search security audit report (audit version 5)" version: "1.2.0" type: report authors: - name: "xixi048" date-released: "2026-10-29" url: "https://skillstore.io/skills/xixi048-feeling-search/audits/5" identifiers: - type: other value: "skillstore:xixi048-feeling-search:audit:5" description: "Skillstore immutable audit report identifier"

Skillstore-Score

Warum dieser Score Evidenzvertrauen: Mittel
64
Architektur
100
Wartbarkeit
87
Inhalt
65
Gemeinschaft
91
Spezifikationskonformität

Was Sie erstellen können

Explore Concept Art Directions

Compare candidate styles and matched mood terms before choosing a visual direction for a scene.

Refine Image Prompt Vocabulary

Look up style terms and related words to make image-generation prompts more specific.

Prepare Brand Reference Notes

Organize style definitions and image provenance into reference notes for a visual identity discussion.

Diese Prompts ausprobieren

Find a Starting Style
Use feeling-search offline for this mood: [description under 500 characters]. Return three candidate styles, definitions, and matched words. Report a missing cache without networking.
Expand Style Vocabulary
Look up [mood word] online and list its associated styles, related words, and opposites. Use endpoints documented as free.
Compare Reference Images
Compare [up to five style identifiers]. Ask before potentially billed requests. Include definitions, image kinds, credits, licenses, and source links. Exclude placeholders.
Run a Controlled Style Search
Analyze [description] locally, then try fuzzy matching if needed. Ask before cloud fallback and confirm possible costs. Return ranked styles, matched terms, source, and limitations.

Bewährte Praktiken

  • Use short, concrete descriptions and review matched words before accepting a style ranking.
  • Download the dictionary with approval, verify cache freshness, and use offline mode when descriptions must stay local.
  • Confirm online costs, protect quota credentials, and preserve image credits and licensing information.

Vermeiden

  • Treating a local lexical match as full semantic understanding or the only correct artistic direction.
  • Enabling cloud analysis, spending quota, or executing the external MCP installer without informed approval.
  • Using placeholder images as authentic examples or assuming all references permit commercial reuse.

Häufig gestellte Fragen

Which AI tools can use this skill?
Claude, Codex, and Claude Code are supported. The bundled client requires Python 3.8 or later and uses standard-library modules.
Does the skill generate images?
No. It returns style candidates, vocabulary, and online reference-image metadata that can inform a separate image-generation workflow.
Can I work entirely offline?
Local analysis needs a previously downloaded, usable dictionary cache. Expired caches can fail offline; word relationships and image lookups require network access.
Is cloud analysis free?
Documentation describes free service, but billing can resume. The bundled client still requires explicit permission and a quota token for cloud fallback.
Does the bundled client understand negation and semantic meaning?
Local analysis uses term matching and optional synonym glosses. Semantic-engine and negation claims refer to an external MCP package or service.
What data can leave my machine?
Cloud analysis sends your description and quota token. Other online commands send lookup terms, style identifiers, credentials, or supplied claim information.

Entwicklerdetails

Autor

xixi048

Lizenz

MIT

Autorenversion

v1.2.0

Skillstore-Revision

r5

Ref.

5b5bad298d69ae8fd37204d293d395c5f80ce32f

Aktualität der Wartung

30.10.2026

Nutzung

3 Downloads · 1 Aufrufe

Dateistruktur

Mehr von xixi048

Alle anzeigen
Alle anzeigen