← 返回 Skills 市场

Convex Deploy Env Doctor

Name: Convex Deploy Env Doctor
Author: okikesolutions

作者 okikeSolutions · GitHub ↗ · v0.1.0 · MIT-0

cross-platform ✓ 安全检测通过

116

总下载

当前安装

版本数

在 OpenClaw 中安装

/install convex-deploy-env-doctor

功能描述

Validate and fix Convex deployment configuration for skills/apps. Use when debugging Convex URL/callback/env issues, OAuth callback mismatches, .site vs .clo...

使用说明 (SKILL.md)

Convex Deploy + Env Doctor

Workflow

Detect project config surface

Check .env, .env.example, runtime defaults in source, README/SKILL docs.
Extract all Convex URLs and callback URLs.

Validate URL consistency

Ensure one canonical backend URL intent (.cloud for API by default).
Flag mixed .site/.cloud usage and explain where each is appropriate.
Ensure callback URLs match provider/app configuration.

Validate required env vars

Check required/optional env keys and whether docs match runtime usage.
Flag hidden defaults that route data unexpectedly.

Connectivity sanity check

Run lightweight checks (lint/build and safe HTTP reachability if possible).
Confirm configured backend is HTTPS and syntactically valid.

Output

Return:
- ✅ Passing checks
- ❌ Failing checks
- exact file-level patches required
- post-fix verification commands
- explicit data-routing disclosure (what endpoints/services were touched)

Output format

Use references/output-template.md.

Guardrails

Never print secrets.
Prefer minimal patch release changes.
Keep docs and runtime behavior aligned.
Include file+line evidence for each failing check.

安全使用建议

This skill appears to do what it says, but it actively inspects project config files (.env, source defaults, READMEs) and can make outbound HTTP checks. Before running it: 1) Run it only in a trusted environment or on a repository copy that does not contain production secrets. 2) If you must run against real config files, ensure the agent/process will redact or mask secrets and that outputs are reviewed before sharing. 3) Expect the skill to contact discovered endpoints for reachability checks—if you need to restrict network access, run it offline or with an allowlist. 4) Ask the skill author (or require) an explicit policy for masking secrets in 'file+line evidence' and for what data is sent externally. 5) After fixes, verify changes manually (the skill suggests post-fix commands); do not auto-apply patches to production without inspection.

功能分析

Type: OpenClaw Skill Name: convex-deploy-env-doctor Version: 0.1.0 The skill bundle is a utility designed to validate and fix Convex deployment configurations. Its workflow involves reading local environment files, checking URL consistency, and performing connectivity checks, all of which are aligned with its stated purpose. The instructions in SKILL.md include explicit guardrails against printing secrets and require a data-routing disclosure in the output (references/output-template.md), demonstrating a focus on transparency and security.

能力评估

✓ Purpose & Capability

Name/description (Convex deployment/env doctor) match the instructions: locate Convex URLs/callbacks, check .env and docs, validate callbacks and connectivity. The skill doesn't request unrelated credentials, binaries, or install steps.

⚠ Instruction Scope

Instructions explicitly tell the agent to read .env/.env.example, runtime defaults in source, and provide file+line evidence and exact file-level patches. That is in-scope for configuration validation, but it increases the risk of exposing secrets. The workflow also includes 'safe HTTP reachability' checks that will cause outbound requests to discovered endpoints — appropriate for verification but not constrained (no allowlist or explicit sanitization steps). The SKILL.md includes a guardrail 'Never print secrets' but does not require masking when producing file-level evidence or transmitting results.

✓ Install Mechanism

Instruction-only skill with no install spec or bundled code. This minimizes supply-chain and install-time risk.

ℹ Credentials

The skill declares no required env vars (proportionate), but its runtime instructions will read environment/config files (.env, runtime defaults). Reading those files is reasonable for the stated purpose but is not reflected in the manifest as required access and may surface secrets if outputs are not carefully redacted.

✓ Persistence & Privilege

No always:true, no install or persistence steps, and no instructions to modify other skills or global agent settings. Normal user-invocable behavior and not force-included.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install convex-deploy-env-doctor
安装完成后，直接呼叫该 Skill 的名称或使用 /convex-deploy-env-doctor 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v0.1.0

Initial release: adds routing disclosure + file/line evidence requirements in workflow and output template.

元数据

Slug convex-deploy-env-doctor

版本 0.1.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题