Engineer production-grade code with AI using battle-tested workflows
Compact the current conversation into a redacted handoff document, with suggested skills, for a fresh agent to continue the work.
Maintainer of this project? Claim this page to edit the listing.
1.0.1Add to Favorites
Why it matters
Enable developers to build real applications with AI agents by providing composable, engineering-focused skills that enforce best practices like alignment through grilling sessions, shared language documentation, test-driven development, and continuous architecture improvement to prevent code entropy.
Outcomes
What it gets done
Align on requirements before coding by running grilling sessions that ask detailed questions and build shared language documentation
Enforce test-driven development with red-green-refactor loops that produce working, well-tested code
Debug systematically using structured diagnostic workflows that identify root causes efficiently
Maintain clean architecture by regularly auditing module design and preventing ball-of-mud codebases
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/ag-handoff | bash Overview
Handoff
Compacts the current conversation into a redacted handoff document with a suggested-skills section, saved to the OS temp directory for a fresh agent to continue the work. Use when ending a session and another agent (or a future session) needs to resume the work with full context.
What it does
Handoff compacts the current conversation into a handoff document so a fresh agent can pick up the work with full context, saved to the user's OS temporary directory rather than the current workspace. It includes a "suggested skills" section naming which skills the next agent should invoke, and if the user passed arguments when invoking it, those are treated as a description of what the next session should focus on and the document is tailored accordingly.
When to use - and when NOT to
Use it when the request is to compact the current conversation into a handoff document for another agent to continue. It is explicitly not meant to duplicate content already captured in other artifacts - PRDs, plans, ADRs, issues, commits, diffs - which should be referenced by path or URL instead of copied into the handoff document.
Inputs and outputs
Input is the conversation so far, plus any arguments describing the next session's focus. Output is a handoff document written to the OS temp directory, with sensitive information - API keys, passwords, personally identifiable information - redacted before it's written.
Integrations
It cross-references other artifact types (PRDs, plans, ADRs, issues, commits, diffs) by path or URL rather than re-embedding their content, and names specific skills the next agent session should invoke.
Who it's for
Agents and users ending a work session who need the next agent (or their own future session) to resume with full context, a redacted and de-duplicated summary, and a clear pointer to which skills to use next.
Source README
When to Use
Use when this workflow matches the user request: Compact the current conversation into a handoff document for another agent to pick up.
_Source: mattpocock/skills (MIT)._Write a handoff document summarising the current conversation so a fresh agent can continue the work. Save to the temporary directory of the user's OS - not the current workspace.
Include a "suggested skills" section in the document, which suggests skills that the agent should invoke.
Do not duplicate content already captured in other artifacts (PRDs, plans, ADRs, issues, commits, diffs). Reference them by path or URL instead.
Redact any sensitive information, such as API keys, passwords, or personally identifiable information.
If the user passed arguments, treat them as a description of what the next session will focus on and tailor the doc accordingly.
Limitations
- Requires the upstream tool, account, API key, or local setup when the workflow names one.
- Does not authorize destructive, production, paid, or external-message actions without explicit user approval.
- Validate generated artifacts or recommendations against the user's real sources before treating them as final.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.