aim-handoff
Ephemeral context handoff: structure Engineering Handoff docs to the gold-standard numbered template (HANDOFF style), update in place, optional Tmux baton…
Skill by BrianV1981
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/brianv1981-aim-handoff
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/aim-handoff
What's inside
- Skill name
aim-handoff- Description
- Ephemeral context handoff: structure Engineering Handoff docs to the gold-standard numbered template (HANDOFF style), update in place, optional Tmux baton pass to a fresh vessel session.
- License
- MIT
- Instructions
- 178 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# `aim-handoff` > **MANDATE:** Execute a one-command end-of-session closeout: automatically verify and ingest any unlogged session work into `memory-wiki/` (Step 1), surgically update the standardized Engineering Handoff document (`HANDOFF.md`), commit & push both directly to `origin/main` with zero confirmation prompts, and optionally pass the baton into a fresh agent via tmux. ## Relationship to other skills | Skill | Role | |-------|------| | **`aim-memory-wiki`** | **Mid-session "Save Game"** — fast (~15s) autonomous checkpoint into `memory-wiki/` without ending the session | | **`aim-handoff`** (this) | **End-of-session closeout** — auto-includes any `memory-wiki/` delta + updates `HANDOFF.md` + pushes both to `origin/main` | | **`aim-precompact-handoff`** | **Safeguard** — Grok context-window guard; forces a handoff *before* auto-compact destroys context | Do not confuse precompact (when/why to write under token pressure) with this skill (what the handoff document must look like). --- ## 0. Step 1 — Idempotent Auto-Wiki Delta Check (MANDATORY) Before touching `HANDOFF.md`, you MUST inspect `memory-wiki/log.md`: 1. **If work or architectural decisions occurred since the last `/aim-memory-wiki`** (or `/aim-memory-wiki` was not run during this session): automatically synthesize the session's discoveries, decisions, and milestones into `memory-wiki/index.md`, `memory-wiki/log.md`, and `memory-wiki/pages/*.md` (and run `./aim index` if LanceDB is active) as Step 1 of this handoff. 2. **If `memory-wiki/` is already current for the session:** skip straight to updating `HANDOFF.md`. --- ## 1. Core philosophy 1. **UPDATE, do not decapitate.** If `HANDOFF.md` exists, you must **surgically update** it. Preserve overarching context. When told to "consolidate completed work", this means appending YOUR new accomplishments into Section 0. It is STRICTLY FORBIDDEN to delete, prune, or compress older historical entries from previous agents. 2. **Wiki first.** Long-term architecture and decisions belong in `memory-wiki/` via `aim-memory-wiki`. The handoff is the relay baton, not a second wiki. 3. **No guesses.** Unknown → write `UNKNOWN`. Proven → path, SHA, issue #, command, count. 4. **No forced architecture.** Do not invent `docs/` trees or roadmaps the Operator did not provide. 5. **Surgical Edits Required:** You MUST use surgical line-by-line file edit tools (e.g. `replace_file_content`) to update the handoff. Generating the entire file from scratch (e.g. using `write_to_file` with Overwrite) is strictly forbidden as it leads to accidental data loss. --- ## 2. Gold-standard template (numbered sections) Canonical shape: gold-standard Engineering Handoff with numbered sections **0–9** (LeadDeed-style `HANDOFF.md` when that project uses it; otherwise `HANDOFF.md`). When instructed to prepare a handoff, generate or **update** a structured markdown file (e.g. `HANDOFF.md` or project-specific `HANDOFF.md`) with **all** of the following sections. Mission-specific titles may specialize (e.g. “MODULE-BY-MODULE ANALYSIS FOR #173”) but the **numbers and roles must remain**: ```markdown # [Project Name] — Engineering Handoff > **Updated:** [Timestamp] > **Updated by:** [Session / vessel id] > **Priority Mission:** [One-line goal] > **Operator:** Brian Vasquez (`@BrianV1981` / `brianv1981@gmail.com` — or project-specific: `bvasquez@leaddeeds.com`, `bvasquez@ayrianna.com`, `bvasquez@skorbane.com`; never invent an email) --- ## 0. COMPLETED WORK (DO NOT REVISIT) | Session | Work | Status | |---------|------|--------| | [id] | [brief] | ✅ RESOLVED | *(Keep clean and consolidated. Point to wiki/issues for deep history — do not re-audit.)* --- ## 1. PROJECT IDENTITY *(1–3 sentences: what the system is, prod vs local, monorepo root. Bullet key envs/URLs.)* ### Your Knowledge Base *(Absolute paths to memory-wiki index + 3–5 critical pages/docs.)* --- ## 2. YOUR MISSION: [TITLE] *(Overarching goal for the *incoming* session.)* ### Execution Queue (in order) #### 1️⃣ [Task] **Problem:** … **Fix:** … **Key files:** [absolute paths] --- ## 3. DETAILED ANALYSIS / BREAKDOWN *(Mission-specific depth: module-by-module, system-by-system, or risk-ordered breakdown. Expand freely; this is where complexity lives so §2 stays scannable.)* --- ## 4. IMPLEMENTATION STRATEGY *(Ordered approach, tiers, pass structure, tests expected. How to attack §2/§3 without thrash.)* --- ## 5. THE CRITICAL TRAPS & WARNINGS > **⚠️ EPISTEMIC / OPERATIONAL WARNINGS** *(Top landmines — e.g. Vercel vs local, wrong env files, destructive git habits.)* --- ## 6. KEY PATHS *(5–15 absolute paths grouped by concern: infra, targets of the mission, config.)* --- ## 7. THE FULL PICTURE / WHAT COMES AFTER *(Dependency graph or short roadmap after the priority mission. Parallel quick wins table OK.)* --- ## 8. OPERATOR PREFERENCES *(Boundaries: do/don't, proof standards, multi-agent coordination, push discipline.)* --- ## 9. IMMEDIATE NEXT STEPS 1. … 2. … *(Imperative, ordered, first command the incoming agent should run.)* ```
From BrianV1981/aim-skill-library, MIT.
More from BrianV1981/aim-skill-library
- 0.601score
aim-memory-wiki
BrianV1981
Interactive active-memory ingestion skill for maintaining a persistent, compounding LLM knowledge base (the Memory Wiki).
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
aim-gitops
BrianV1981
Automates the strict GitOps worktree and TDD workflow (formerly `aim fix` and `aim promote`) natively using PowerShell and git.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
aim-mega-guide
BrianV1981
A highly specialized content pipeline for generating consistent, deeply structured 'Mega Guides' and SOPs from a single, long-form YouTube video or podcast…
SkillAFreshness: A★ 3Claude Code +9 - 0.597score
aim-mega-article
BrianV1981
Multi-source B2B authority article pipeline combining 3-5 contextual YouTube masterclass embeds, inline hyperlinked empirical data citations, 3 bespoke…
SkillAFreshness: A★ 3Claude Code +9 - 0.569score
aim-communicate
BrianV1981
Inter-agent communication via tmux with permission gates, chalkboard + short-paste protocol (Grok/AGY-hardened).
SkillBFreshness: B★ 3Claude Code +9 - 0.566score
aim-audit
BrianV1981
Evolving product-truth audit: one living artifacts/AUDIT_<SLUG>.md, update in place (aim-handoff rules), never dated _REAUDIT/_AUDIT3 novels.
SkillBFreshness: B★ 3Claude Code +9 - 0.566score
aim-calc
BrianV1981
Stateful scientific calculator for agents.
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
aim-grok-context
BrianV1981
Project context for the aim-grok vessel (A.I.M.
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
aim-export-cartridge
BrianV1981
Export A.I.M.
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
aim-bwrap-forge
BrianV1981
Dynamically spin up bubblewrap (bwrap) isolated sandbox directories and optional tmux co-agent sessions for untrusted or multi-tenant work.
SkillBFreshness: B★ 3Claude Code +9 - 0.558score
aim-projects
BrianV1981
Interacts with the GitHub Projects kanban board natively using the `gh` CLI on Windows.
SkillBFreshness: B★ 3Claude Code +9 - 0.558score
aim-google
BrianV1981
Fast, script-friendly CLI for interacting with Google Workspace (Gmail, Calendar, Drive, Docs, etc.)
SkillBFreshness: B★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 0
- Open issues
- 6
- Last commit (this folder)
- Oct 22, 2026 (6 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/BrianV1981/aim-skill-library
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
More from BrianV1981
- 0.601score
aim-memory-wiki
BrianV1981
Interactive active-memory ingestion skill for maintaining a persistent, compounding LLM knowledge base (the Memory Wiki).
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
aim-gitops
BrianV1981
Automates the strict GitOps worktree and TDD workflow (formerly `aim fix` and `aim promote`) natively using PowerShell and git.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
aim-mega-guide
BrianV1981
A highly specialized content pipeline for generating consistent, deeply structured 'Mega Guides' and SOPs from a single, long-form YouTube video or podcast…
SkillAFreshness: A★ 3Claude Code +9 - 0.597score
aim-mega-article
BrianV1981
Multi-source B2B authority article pipeline combining 3-5 contextual YouTube masterclass embeds, inline hyperlinked empirical data citations, 3 bespoke…
SkillAFreshness: A★ 3Claude Code +9 - 0.569score
aim-communicate
BrianV1981
Inter-agent communication via tmux with permission gates, chalkboard + short-paste protocol (Grok/AGY-hardened).
SkillBFreshness: B★ 3Claude Code +9 - 0.566score
aim-audit
BrianV1981
Evolving product-truth audit: one living artifacts/AUDIT_<SLUG>.md, update in place (aim-handoff rules), never dated _REAUDIT/_AUDIT3 novels.
SkillBFreshness: B★ 3Claude Code +9
Cite this page
aim-handoff skill. Agent Mart. https://agentmart.dev/skills/brianv1981-aim-handoff (last updated 2026-10-24)