跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
浏览器与自动化

feedback

Submit feedback, bug reports, feature requests, or share SDLC patterns you've discovered. Privacy-first — always asks before scanning.

BaseInfinity · 已有 AI 审核记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add BaseInfinity/claude-sdlc-harness --skill feedback

技能说明


name: feedback description: Submit feedback, bug reports, feature requests, or share SDLC patterns you've discovered. Privacy-first — always asks before scanning. argument-hint: "[optional: bug | feature | pattern | improvement]" effort: medium

Feedback — Community Contribution Loop

Task

$ARGUMENTS

Purpose

Help users contribute back to the SDLC wizard: bug reports, feature requests, pattern sharing, and SDLC improvements. Privacy-first — never scan without explicit permission.

Privacy & Permission (MANDATORY)

NEVER scan the user's repo without explicit consent. Always ask first:

"I can scan your SDLC setup to identify what you've customized vs wizard defaults. This helps me create a more specific report. May I scan? (Only file names and SDLC config are read — no source code, secrets, or business logic.)"

What IS scanned (with permission):

  • SDLC.md, TESTING.md, CLAUDE.md structure (not content details)
  • Hook file names and which hooks are active
  • Skill names and which skills exist
  • .claude/settings.json hook configuration (not allowedTools or secrets)

What is NEVER scanned:

  • Source code files
  • .env files, secrets, credentials
  • Business logic or proprietary code
  • Git history or commit messages

Feedback Types

Bug Report

  1. Ask user to describe the issue
  2. With permission, check which wizard version is installed (SDLC.md metadata)
  3. Check if hooks are properly configured
  4. Create a GitHub issue with reproduction steps

Feature Request

  1. Ask user what they want
  2. With permission, check if a similar capability already exists in their setup
  3. Create a GitHub issue with the request and context

Pattern Sharing

  1. Ask user what pattern they've discovered (custom hook, modified philosophy, test approach)
  2. With permission, diff their SDLC setup against wizard defaults to identify customizations
  3. Ask: "Which of these customizations worked well for you?"
  4. Create a GitHub issue describing the pattern and evidence it works

SDLC Improvement

  1. Ask what could be better about the SDLC workflow
  2. With permission, check which SDLC steps they use most/least
  3. Create a GitHub issue with the improvement suggestion

Creating the Issue

Use gh issue create on the wizard repo:

gh issue create \
  --repo BaseInfinity/claude-sdlc-harness \
  --title "[feedback-type]: Brief description" \
  --body "$(cat <<'EOF'
## Feedback Type
bug / feature / pattern / improvement

## Description
[User's description]

## Context
- Wizard version: [from SDLC.md metadata]
- Setup type: [detected stack if permission granted]

## Evidence (if pattern sharing)
[What the user customized and why it worked]

---
Submitted via `/feedback` skill
EOF
)"

Rules

  • Privacy first — always ask before scanning anything
  • Opt-in only — if user declines scan, still create the issue with whatever they tell you manually
  • No source code — never include source code snippets in issues
  • Be specific — vague issues waste maintainer time. Ask clarifying questions
  • Check for duplicates — gh issue list --repo BaseInfinity/claude-sdlc-harness --search "keywords" before creating