Token导航 LogoToken导航

Agent Skill · 研究检索

feedback-session(反馈会议)

feedback-session 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

feedback-session 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为执行命令;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 46,073。

总安装 808周安装 33GitHub Stars 28下载 259核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:feedback-session(反馈会议)
来源仓库:https://github.com/laurigates/claude-plugins
仓库路径:skills/feedback-session
安装命令:
npx skills add https://github.com/laurigates/claude-plugins --skill feedback-session
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/laurigates/claude-plugins --skill feedback-session

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

反馈会议技能用于分析当前会话中的技能使用反馈。可创建GitHub Issue跟踪Bug、改进点和积极实践模式。适合在发现技能指引错误或需要静态质量分析时使用。安装前需确认是否会触发命令执行或文件写入操作。建议结合项目上下文理解反馈的具体应用场景。

feedback-session 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/feedback-session

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/laurigates/claude-plugins --skill feedback-ses…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要执行命令,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

feedback-session 怎么安装?

在终端执行 npx skills add https://github.com/laurigates/claude-plugins --skill feedback-session。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

feedback-session 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

feedback-session 需要什么权限?

权限级别是执行命令。安装流程涉及命令执行,可能通过 npx skills add https://github.com/laurigates/claude-plugins --skill feedback-session 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

feedback-session 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

feedback-session 用的人多吗?

公开数据是累计安装 808 次,近一周安装 33 次,下载 259 次,GitHub 28 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

feedback-session 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要执行命令,装到常用环境前要先评估影响面。
  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 不能接受 Skill 在你机器上执行命令的人。
  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
feedback-session

反馈会议 · 本页

80828执行命令未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

/feedback:session

Analyze the current session for skill feedback and create GitHub issues to track bugs, enhancements, and positive patterns.

When to Use This Skill

Use this skill when...Use alternative when...
A skill gave wrong or outdated guidanceWant to update skills directly -> /project:distill
A command failed due to skill adviceNeed static skill quality analysis -> /health:audit
Discovered a better flag or patternWant to capture general learnings -> /project:distill
A skill worked particularly wellWant to track command usage stats -> /analytics-report
End of session, want to file feedbackNeed to fix a skill right now -> edit the SKILL.md directly
Feedback is about the plugin itselfUse --target-repo laurigates/claude-plugins to file against the plugin source

Known Limitations

IaC-managed labels: Some repositories manage GitHub labels declaratively via Terraform, Pulumi, or similar tools. In these repos, gh label create will either be forbidden or cause drift that the IaC tool destroys on the next apply. This skill detects this case and offers a graceful fallback (see Step 1).

Default target repo: By default, this skill files issues against the repository in the current working directory. If you are giving feedback about a plugin skill itself rather than the application code in the session, use --target-repo <owner/repo> to point at the plugin source repo.

Context

  • Git remotes:!git remote -v
  • Open feedback issues:!gh issue list --label session-feedback --state open --json number,title --jq '.[].title'
  • Open positive issues:!gh issue list --label positive-feedback --state open --json number,title --jq '.[].title'

Parameters

Parse these from $ARGUMENTS:

ParameterDescription
--dry-runShow findings without creating issues
--bugs-onlyOnly report bugs (wrong/outdated guidance)
--enhancements-onlyOnly report enhancement opportunities
--positive-onlyOnly report positive feedback
--target-repo <owner/repo>File issues against this repo instead of the cwd repo
-R <owner/repo>Alias for --target-repo
[plugin-name]Scope analysis to a specific plugin

After parsing, set $TARGET_REPO to the value of --target-repo/-R if provided. Append -R $TARGET_REPO to all gh commands below when $TARGET_REPO is set.

Execution

Execute this session feedback workflow:

Step 1: Resolve target repo and ensure labels exist

1a. Determine target repo

If --target-repo or -R was passed in $ARGUMENTS, set $TARGET_REPO to that value and append -R $TARGET_REPO to every gh command in the remaining steps.

If not provided, infer the repo from the cwd: gh repo view --json nameWithOwner -q '.nameWithOwner'. Use this as the implicit target (no -R flag needed since gh defaults to cwd).

1b. Check whether labels are IaC-managed

Run: gh label list -R $TARGET_REPO --json name,description --jq '.[].description' (omit -R if no explicit target).

Scan the output for IaC indicators in any label description:

  • Keywords: terraform, pulumi, cdk, managed by, do not create, iac, infrastructure

Also check for labels.tf in the cwd: look for files matching **/labels.tf or **/labels.yaml patterns using Glob.

If IaC indicators are found or labels.tf / labels.yaml exist in the working tree:

  • Display a warning: ` ⚠ IaC-managed labels detected in <repo>. The session-feedback and positive-feedback labels cannot be created via gh label create — they are managed declaratively and creating them out-of-band would cause drift. `
  • Use AskUserQuestion to ask: How would you like to proceed? Options: If user chooses option 2, set $TARGET_REPO to their input and re-run step 1b for the new repo. If user chooses option 3, exit. If user chooses option 1, set $SKIP_SESSION_LABELS=true and continue.

1. Proceed without session-feedback labels — issues will be created with only bug/enhancement labels; add the two labels to your IaC definition to backfill. 2. Use a different target repo — enter an owner/repo where you can create labels freely (e.g. laurigates/claude-plugins). 3. Abort — stop here.

1c. Create missing labels (only when not IaC-managed)

Skip this step if $SKIP_SESSION_LABELS=true.

  1. Check if session-feedback exists: gh label list --json name --jq '.[].name' | grep -q session-feedback
  2. If missing: gh label create session-feedback --description "Feedback from session analysis" --color "d876e3"
  3. Check if positive-feedback exists similarly.
  4. If missing: gh label create positive-feedback --description "Skills that worked well" --color "0e8a16"

Step 2: Analyze conversation history

Review the entire conversation for feedback signals. Look for these categories:

Bugs (label: session-feedback, bug):

  • Skill gave wrong command syntax or outdated flags
  • Command failed because skill guidance was incorrect
  • Skill recommended a pattern that caused errors
  • Skill was missing a critical caveat or prerequisite

Enhancements (label: session-feedback, enhancement):

  • Discovered a better flag or option than what the skill suggests
  • Found a workflow gap the skill should cover
  • Identified a missing pattern or integration
  • Found a more efficient approach than the skill recommends

Positive (label: positive-feedback):

  • Skill provided correct, effective guidance
  • Skill's agentic optimizations saved time
  • Skill's decision table correctly directed to the right tool
  • Skill's patterns worked well in practice

For each finding, record:

  • Category: bug, enhancement, or positive
  • Plugin: which plugin the skill belongs to
  • Skill: which specific skill
  • Description: what happened
  • Evidence: the specific interaction or error that demonstrates it

Filter by $ARGUMENTS:

  • If --bugs-only: only report bugs
  • If --enhancements-only: only report enhancements
  • If --positive-only: only report positive feedback
  • If [plugin-name] specified: only report for that plugin

Step 3: Deduplicate against open issues

For each finding, search for existing issues in $TARGET_REPO:

gh issue list --label session-feedback --search "<skill-name> <key-phrase>" --json number,title --jq '.[].title'

Skip findings that match an existing open issue title. Note skipped items for the summary.

If $SKIP_SESSION_LABELS=true, search without labels: gh issue list --search "feedback(<plugin>)" --json number,title --jq '.[].title'

Step 4: Present findings for review

Use AskUserQuestion to present categorized findings. Group by category:

Format each finding as:

[BUG] plugin-name/skill-name: brief description
[ENH] plugin-name/skill-name: brief description
[POS] plugin-name/skill-name: brief description

Let the user select which findings to file as issues (use multiSelect).

If --dry-run, present findings and stop here.

Step 5: Create approved issues

For each approved finding, create a GitHub issue in $TARGET_REPO:

Title format: feedback(<plugin-name>): <description>

Labels (when not $SKIP_SESSION_LABELS):

  • Bugs: session-feedback, bug
  • Enhancements: session-feedback, enhancement
  • Positive: positive-feedback

Labels (when $SKIP_SESSION_LABELS=true):

  • Bugs: bug
  • Enhancements: enhancement
  • Positive: *(no label — omit the --label flag)*

Body template:

## Skill

`<plugin-name>/skills/<skill-name>/SKILL.md`

## Category

<Bug | Enhancement | Positive feedback>

## Description

<What happened during the session>

## Evidence

<Specific interaction, error message, or successful outcome>

## Suggested Action

<What should change in the skill, or what should be preserved>

Create each issue:

gh issue create --title "feedback(<plugin>): <desc>" --label "<labels>" --body "<body>"

Append -R $TARGET_REPO when set. Omit --label if no labels apply (positive + $SKIP_SESSION_LABELS).

Step 6: Report summary

Print a summary:

MetricCount
Findings identifiedN
Duplicates skippedN
Issues createdN
Skipped by userN

List created issue numbers with links. If $SKIP_SESSION_LABELS=true, remind the user to add session-feedback and positive-feedback to their IaC label definition.

Agentic Optimizations

ContextCommand
List feedback issuesgh issue list --label session-feedback --json number,title,labels -q '.[]'
Search for duplicatesgh issue list --label session-feedback --search "keyword" --json title -q '.[].title'
Detect IaC label signalsgh label list --json name,description --jq '.[].description'
Check label existsgh label list --json name -q '.[].name'
Create labelgh label create name --description "desc" --color "hex"
Create issue (with target)gh issue create -R owner/repo --title "t" --label "l1,l2" --body "b"
Create issue (no labels)gh issue create -R owner/repo --title "t" --body "b"
Infer current repogh repo view --json nameWithOwner -q '.nameWithOwner'

Quick Reference

FlagDescription
--dry-runShow findings without creating issues
--bugs-onlyOnly bug reports
--enhancements-onlyOnly enhancement suggestions
--positive-onlyOnly positive feedback
--target-repo <owner/repo>File issues against a different repo (e.g. plugin source)
-R <owner/repo>Alias for --target-repo
[plugin-name]Scope to specific plugin

平台分布

Codex

36.72%
按下载量换算95

Claude

28.89%
按下载量换算75

Cursor

19.25%
按下载量换算50

Gemini CLI

9.64%
按下载量换算25

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/laurigates/claude-plugins --skill feedback-session 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills