OpenAgentSkill
GitHub200+

Évaluation avant installation

automating-devops Rapport d’évaluation.

Une décision d’installation lisible par machine pour les Agents : adéquation à la tâche, Trust Score, Audit Score, sécurité d’installation, surface de permissions et plan de validation concret avant de toucher l’espace de travail.

Revue nécessaireRisque moyenRevoir Politique
68
Rapport d’évaluation
69
Confiance
74
Audit
54
Sécurité Agent

Revue manuelle

Test manually in an isolated workspace and compare against safer alternatives.

Gates requis

Vérifications qu’un Agent doit réussir avant installation

Ouvrir JSON

Adéquation à la tâche

84

Validé

Task wording matches this skill metadata.

  • Evaluate automating-devops before installing it in an agent workflow
  • Design et création
  • GitHub automation workflows; Claude Code teams; builders willing to evaluate younger projects

Chemin d’installation

92

Validé

Install handoff is available.

  • npx skills add telagod/code-abyss --skill automating-devops

Sécurité de la commande d’installation

92

Validé

Chemin d’installation standard de package ou runtime

  • npx skills add telagod/code-abyss --skill automating-devops

Score de confiance

69

Avertissement

Potentially useful, but at least one trust signal needs human inspection.

  • Revue manuelle
  • 239 stars GitHub
  • MIT

Score d’audit

74

Avertissement

Revue nécessaire

  • SKILL.md references `skills/_kernel/backend/SKILL.md` which is not part of this skill directory; this creates a dependency on an external file that may not be available when the skill is used in isolation.

Contrôle de sécurité Agent

54

Avertissement

Sparse or mixed signals. Useful for discovery, but not for autonomous installation.

  • Test manually in an isolated workspace and compare against safer alternatives.
  • SKILL.md references `skills/_kernel/backend/SKILL.md` which is not part of this skill directory; this creates a dependency on an external file that may not be available when the skill is used in isolation.

Clarté de licence

86

Validé

MIT

  • MIT

Surface de permissions

74

Avertissement

network or browser access, database access

  • Browser automation: medium
  • Network access: medium
  • Database access: medium

Plan de validation

Ce que l’Agent doit faire ensuite

  1. 1Inspect repository, README/SKILL.md, license, and recent commits before production use.
  2. 2Install in an isolated workspace or sandbox with no production secrets available.
  3. 3Run the smallest representative task and record files touched, commands run, network access, and outputs.
  4. 4Compare the selected skill against at least one alternative when the eval status is review or failed.
  5. 5Promote only after the agent reports a successful verification result and unresolved warnings are accepted.

Ne pas utiliser quand

Conditions exigeant une autre skill

  • Équipes qui nécessitent un SLA soutenu par le fournisseur
  • production agents without a repository review
  • SKILL.md references `skills/_kernel/backend/SKILL.md` which is not part of this skill directory; this creates a dependency on an external file that may not be available when the skill is used in isolation.
  • The skill is primarily in Chinese, which may limit accessibility for non-Chinese-speaking agents; the description is in English but the content is not.
  • Quality score needs review
  • Stars/forks activity: 239 stars, 32 forks; issue activity unavailable in current metadata

Vérifications complémentaires

Signaux de confiance derrière la décision

Complétude README/SKILL.md

Validé

86

Les métadonnées incluent suffisamment de contexte d’usage et de workflow

Maintenance récente

Validé

88

3 mois depuis le dernier push

Alternatives disponibles

Validé

82

Alternative skills are available for comparison.