skills.rest
ThorStarlord
ThorStarlordCOMMUNITY
Explore all skills by this author →
GitHub

handoff

Generate a Git- and pytest-based handoff document with decisions and next task.

1|Updated May 6, 2026
Search Tags:#documentation#pytest#handoff#ADR#architectural decisions#agent collaboration#git log
One-click install
npx skills add https://github.com/ThorStarlord/interface-skills --skill handoff-thorstarlord
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: handoff
Source: https://github.com/ThorStarlord/interface-skills/tree/main/skills/handoff
Command: npx skills add https://github.com/ThorStarlord/interface-skills --skill handoff-thorstarlord

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Handoffs fail when the next agent or session lacks an accurate, evidence-based picture of what changed, what was decided, and what to do next.

Core Features & Use Cases

  • Evidence-first session recap: Builds a feature completion summary from recent git history, diffs, conversation context, and final test results to reduce guesswork and spec drift.
  • Actionable next-step frontier: Forces a single concrete next technical task with file/function targets, an approach statement, and the specific test to write.
  • Decision continuity: Extracts architectural decisions and enumerates locked ADRs from both conversation and docs/adr/ to prevent regressions caused by re-arguing choices.

Quick Start

Ask an AI agent to run the handoff workflow and save the summary to docs/handoffs/YYYY-MM-DD-feature-name.md using the provided template.

Frequently Asked Questions about handoff

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I document a feature completion session for another AI agent?▼

Generate a handoff document by compiling git history, diffs, conversation context, and pytest results into a markdown file. This provides an evidence-first session recap for the next agent without replaying context.

How do I transfer architectural decisions to a new agent without losing context?▼

Transfer architectural decisions by extracting them from the conversation and enumerating locked ADRs from docs/adr/ into a handoff document, preventing regressions caused by re-arguing choices.

What should an AI agent handoff document include for software engineering tasks?▼

An AI agent handoff document should include recent commits, modified files, exact pytest counts, architectural decisions with rejected alternatives, locked ADRs, and a single concrete next frontier task.

Can I use pytest results and git log to create a session summary?▼

Yes, you can compile pytest results and git log into a handoff document. The workflow extracts git history, diffs, and exact test counts to build an evidence-based feature completion recap.

What is the best way to define the next technical task when handing off work?▼

The best way to define the next technical task is to specify a single concrete step with file/function targets, an approach statement, and the specific test to write, creating an actionable next-step frontier.

Where should agent handoff documents be saved in a project?▼

Agent handoff documents should be produced as markdown files at docs/handoffs/YYYY-MM-DD-feature-name.md, standardizing session summaries for future context transfer.

Related Skills

Explore More Software Engineering Skills→
pipecat-aipipecat-ai
14.0k

pr-description

Generate concise GitHub PR descriptions from commits and diffs.

Official
Intermediate
dixusdixus
10

continue

Load the latest handoff file from .claude/handoffs and resume work.

Community
Intermediate
simonheimlichersimonheimlicher
1

architecting-typescript

Create Architecture Decision Records for TypeScript architecture decisions.

Community
Advanced
DavidROliverBADavidROliverBA
49

/wipe

Clear the current AI session and generate a context handoff summary.

Community
Intermediate