look-back — for Claude Code look-back, nexo-agent, community, for Claude Code, ide skills, $ARGUMENTS, briefing, home_items[], grocery_items[], working_note_reminders[]

v1.0.0

关于此技能

适用场景: Ideal for AI agents that need look-back briefing. 本地化技能摘要: # Look-back briefing Retrospective review. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

功能特性

Look-back briefing
Caller-supplied window/focus: $ARGUMENTS
Default: since the most recent prior snapshot. That's your baseline.
If Matt names a specific prior briefing ("compare to yesterday's 8am brief"), use that as the
What this skill emphasizes

# 核心主题

ellison-projects ellison-projects
[0]
[0]
更新于: 4/27/2026

Killer-Skills Review

Decision support comes first. Repository text comes second.

Reference-Only Page Review Score: 10/11

This page remains useful for teams, but Killer-Skills treats it as reference material instead of a primary organic landing page.

Original recommendation layer Concrete use-case guidance Explicit limitations and caution Quality floor passed for review
Review Score
10/11
Quality Score
61
Canonical Locale
en
Detected Body Locale
en

适用场景: Ideal for AI agents that need look-back briefing. 本地化技能摘要: # Look-back briefing Retrospective review. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

核心价值

推荐说明: look-back helps agents look-back briefing. Look-back briefing Retrospective review. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

适用 Agent 类型

适用场景: Ideal for AI agents that need look-back briefing.

赋予的主要能力 · look-back

适用任务: Applying Look-back briefing
适用任务: Applying Caller-supplied window/focus: $ARGUMENTS
适用任务: Applying Default: since the most recent prior snapshot. That's your baseline

! 使用限制与门槛

  • 限制说明: In that case, work only from public/briefings/ and don't write or commit anything
  • 限制说明: Only surface them if they help explain drift
  • 限制说明: Requires repository-specific context from the skill documentation

Why this page is reference-only

  • - Current locale does not satisfy the locale-governance contract.

Source Boundary

The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.

评审后的下一步

先决定动作,再继续看上游仓库材料

Killer-Skills 的主价值不应该停在“帮你打开仓库说明”,而是先帮你判断这项技能是否值得安装、是否应该回到可信集合复核,以及是否已经进入工作流落地阶段。

实验室 Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

常见问题与安装步骤

以下问题与步骤与页面结构化数据保持一致,便于搜索引擎理解页面内容。

? FAQ

look-back 是什么?

适用场景: Ideal for AI agents that need look-back briefing. 本地化技能摘要: # Look-back briefing Retrospective review. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

如何安装 look-back?

运行命令:npx killer-skills add ellison-projects/nexo-agent/look-back。支持 Cursor、Windsurf、VS Code、Claude Code 等 19+ IDE/Agent。

look-back 适用于哪些场景?

典型场景包括:适用任务: Applying Look-back briefing、适用任务: Applying Caller-supplied window/focus: $ARGUMENTS、适用任务: Applying Default: since the most recent prior snapshot. That's your baseline。

look-back 支持哪些 IDE 或 Agent?

该技能兼容 Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer。可使用 Killer-Skills CLI 一条命令通用安装。

look-back 有哪些限制?

限制说明: In that case, work only from public/briefings/ and don't write or commit anything;限制说明: Only surface them if they help explain drift;限制说明: Requires repository-specific context from the skill documentation。

安装步骤

  1. 1. 打开终端

    在你的项目目录中打开终端或命令行。

  2. 2. 执行安装命令

    运行:npx killer-skills add ellison-projects/nexo-agent/look-back。CLI 会自动识别 IDE 或 AI Agent 并完成配置。

  3. 3. 开始使用技能

    look-back 已启用,可立即在当前项目中调用。

! 参考页模式

此页面仍可作为安装与查阅参考,但 Killer-Skills 不再把它视为主要可索引落地页。请优先阅读上方评审结论,再决定是否继续查看上游仓库说明。

Upstream Repository Material

The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.

Upstream Source

look-back

# Look-back briefing Retrospective review. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows. Look-back briefing

SKILL.md
Readonly
Upstream Repository Material
The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.
Supporting Evidence

Look-back briefing

Retrospective review. Snapshot-history-first. Celebrates what closed, surfaces what drifted, notices patterns across the window.

Your task

Run the look-back flow below and return the review to the caller. Parse $ARGUMENTS for window selection (e.g. "this week", "since Monday", "compare to yesterday's 8am brief") or emphasis. If empty, default to the window described below (since the most recent prior snapshot).

Caller-supplied window/focus: $ARGUMENTS

Fetch + snapshot

Same mechanics as the briefing skill — see .claude/skills/briefing/SKILL.md for filename format, error handling, storage location, and the commit-and-push step. By default, fetch GET /api/agent/briefing, write a snapshot so "now" is captured and future look-backs have one more data point, then commit and push it per the canonical spec.

Skip the fetch when the user explicitly asks for a pure-historical view ("just compare Monday's snapshot to Wednesday's, don't fetch"). In that case, work only from public/briefings/ and don't write or commit anything.

Window selection

  • Default: since the most recent prior snapshot. That's your baseline.
  • If Matt names a window ("this week", "since Monday", "last 3 days", "since <date>"), pick the earliest snapshot in that window as the baseline and compare against the latest (or the one just written).
  • If Matt names a specific prior briefing ("compare to yesterday's 8am brief"), use that as the baseline.

If the window has no snapshots (or only the one just written), say so: "I only have snapshots from <earliest date> onward — nothing covers that period." Don't guess.

What this skill emphasizes

Diff the baseline snapshot against the current one by id across these fields (all list what's open, so an id disappearing means completed or dismissed):

  • home_items[]
  • grocery_items[]
  • working_note_reminders[] and checked items in working_notes[].items[]
  • ai_reminders.overdue[], ai_reminders.upcoming[]

Categories:

  • Finished: ids in baseline but not in current. Also: working-note items now checked: true that weren't before.
  • Drifted: ids in baseline that are still open now. Count how long — scan snapshots backward to find the id's first appearance (in briefings, days).
  • Patterns: themes visible across the window. Examples: "closed 5 home items, 0 plan items"; "same reminder has slipped three briefings"; "no moments logged this week".
  • What didn't happen: working-note reminders that were flagged at the start of the window and are still flagged now.

De-emphasize: upcoming dates, overdue reminders looking forward — those belong in look-ahead. Only surface them if they help explain drift.

Output shape

Look-back — <window, e.g. "since <prior timestamp>, 18h ago" or "since Monday, 3 days, 4 snapshots">

Finished:
- #<id> <description>  (<category: home / plan / groceries / reminder>)

Drifted (still open):
- #<id> <description>  — open since <first-seen timestamp>, <N> briefings

Patterns:
- <1–3 observations grounded in the data>

Worth reflecting on:
<1–2 sentences. What the window seems to be saying.>

Section rules:

  • Cap each list at 5 items; roll up the rest as +N more.
  • Omit empty sections.
  • If nothing closed and nothing drifted: "No movement in this window."

Voice

Reflective, honest, calm. "You closed five home items this week — most of any window in your history." Or: "That shower regrout has been on the list for eight briefings — worth a decision: do it, delegate it, or drop it." No flattery, no scolding. Observations, not commentary.

Read docs/matt/goals.md if patterns need framing against longer-term priorities. Not by default.

Constraints

  • Read-only. If Matt wants to close something or add a note in response, the nexo-prm skill handles non-person writes and nexo-people handles person-attached writes.
  • Don't call other NexoPRM endpoints to enrich the view — the briefing response plus snapshot history is the full source.
  • Don't manufacture patterns. If the window is too short or the snapshots too sparse to say anything confident, say that instead.
  • Don't log or echo NEXO_API_KEY.

相关技能

寻找 look-back 的替代方案 (Alternative) 或可搭配使用的同类 community Skill?探索以下相关开源技能。

查看全部

openclaw-release-maintainer

Logo of openclaw
openclaw

本地化技能摘要: 🦞 # OpenClaw Release Maintainer Use this skill for release and publish-time workflow. It covers ai, assistant, crustacean workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

333.8k
0
AI

widget-generator

Logo of f
f

本地化技能摘要: Generate customizable widget plugins for the prompts.chat feed system # Widget Generator Skill This skill guides creation of widget plugins for prompts.chat . It covers ai, artificial-intelligence, awesome-list workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf

149.6k
0
AI

flags

Logo of vercel
vercel

本地化技能摘要: The React Framework # Feature Flags Use this skill when adding or changing framework feature flags in Next.js internals. It covers blog, browser, compiler workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

138.4k
0
浏览器

pr-review

Logo of pytorch
pytorch

本地化技能摘要: Usage Modes No Argument If the user invokes /pr-review with no arguments, do not perform a review . It covers autograd, deep-learning, gpu workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

98.6k
0
开发者工具