aftership-ai-layer Eval
=======================

Status: review
Score: 65/100
Risk: medium
Decision: manual_review
Policy: review
Reason: Test manually in an isolated workspace and compare against safer alternatives.

Install:
npx skills add wubin1836/ai-hive-agent-skills --skill aftership-ai-layer

Required checks:
- PASS Task fit: Task wording matches this skill metadata.
- PASS Install path: Install handoff is available.
- PASS Install command safety: standard package or runtime install path
- WARN Trust score: Potentially useful, but at least one trust signal needs human inspection.
- WARN Audit score: Needs review
- WARN Agent safety gate: Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
- PASS License clarity: MIT
- PASS Permission surface: database access

Warnings:
- Trust score: Potentially useful, but at least one trust signal needs human inspection.
- Audit score: Needs review
- Agent safety gate: Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
- README/SKILL.md completeness: Public metadata needs stronger README/SKILL.md context
- SKILL.md describes AfterShip as a '全球ERP、库存、订单与履约' platform, but AfterShip is primarily a post-purchase tracking/shipping experience platform, not a full ERP. This factual mismatch could mislead agents into recommending the wrong system-of-record boundaries.
- The skill depends on an AI-HIVE MCP being available and calls 'tools/list', but it does not document setup prerequisites such as connection details, authentication, environment variables, or fallback behavior when the MCP tools are unavailable.
- There is no explicit prompt-injection guardrail for external or generated content. AI-HIVE outputs and uploaded assets should always be treated as data, not as instructions to execute.
- The JSON contract mentions controls like sandbox and human approval but does not include concrete budget limits, timeouts, or stop conditions, even though budget and rollback are discussed in prose.
- Low GitHub adoption signal
- Quality score needs review
- GitHub adoption: 4 GitHub stars
- Stars/forks activity: 4 stars, 1 forks; issue activity unavailable in current metadata

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

Do not use when:
- teams that need a vendor-supported SLA
- production agents without a repository review
- Low GitHub adoption signal
- SKILL.md describes AfterShip as a '全球ERP、库存、订单与履约' platform, but AfterShip is primarily a post-purchase tracking/shipping experience platform, not a full ERP. This factual mismatch could mislead agents into recommending the wrong system-of-record boundaries.
- No OpenAgentSkill engagement data yet
- The skill depends on an AI-HIVE MCP being available and calls 'tools/list', but it does not document setup prerequisites such as connection details, authentication, environment variables, or fallback behavior when the MCP tools are unavailable.
- There is no explicit prompt-injection guardrail for external or generated content. AI-HIVE outputs and uploaded assets should always be treated as data, not as instructions to execute.
- The JSON contract mentions controls like sandbox and human approval but does not include concrete budget limits, timeouts, or stop conditions, even though budget and rollback are discussed in prose.

URLs:
- Skill: https://www.openagentskill.com/skills/wubin1836-aftership-ai-layer
- Audit: https://www.openagentskill.com/skills/wubin1836-aftership-ai-layer/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=wubin1836-aftership-ai-layer