Menu
Writes a structured session handoff so a fresh agent can continue without losing thread
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
| name | handoff |
| description | Writes a structured session handoff so a fresh agent can continue without losing thread |
Different from /compact (which compresses context):
Type: /handoff [optional: brief description of what was happening]
The agent writes a handoff document to: ~/.hermes/handoffs/YYYY-MM-DD-HH-MM.md
Structure:
# Handoff — [date time AEST]
## What was being worked on
[1 paragraph — specific task, not vague]
## Current state
- What's DONE (committed/verified)
- What's IN PROGRESS (started but not committed)
- What's BLOCKED and why
## Files modified this session
[list with one-line description of change per file]
## The next agent should
[Numbered list — specific actions, not "continue the work"]
## Context the next agent needs
[Key facts, decisions made this session, any gotchas discovered]
## Commands to run first
```bash
# Verify state before continuing
[e.g. pnpm type-check, git status, etc.]
[Anti-patterns or dead ends discovered this session]
## Usage patterns
### Handing off to prototype
"I've drafted the spec, /handoff to prototype"
→ Writes handoff → prototype agent reads it → picks up from spec
### Handing off mid-feature
"/handoff — auth middleware half-done"
→ Writes handoff → fresh agent knows exact stopping point
### Cross-session (close laptop)
"/handoff end of day"
→ Writes handoff → tomorrow's agent reads it at session start
## Integration with soul.md
At session start, Hermes agents check ~/.hermes/handoffs/ for the most recent handoff.
If found (< 24 hours old), inject it alongside soul.md as Tier-0 context.
Evaluates agent output adversarially before committing — prevents self-assessment bias
Turn a plain-English vision or task into a verified, build-ready spec before any code. Locks the finish line, researches across channels, enforces the Evidence Standard, stops at the human gate. Adopted from CleanExpo/Fabel-Prompt-Engineer for the Authority-Site / Pi-CEO / Unite-Group build process.
Loop through advisor profiles in knowledge/board/ and return a combined multi-persona critique of a spec, draft, or decision.
Turn a plain-English vision into a verified, build-ready spec. Locks the finish line, researches across channels, enforces the Evidence Standard, and stops at the human approval gate.
Capture the user's feedback about an output and update the relevant skill or knowledge files so future outputs get sharper.
File new articles, transcripts, and notes into knowledge/ in the right place with source metadata and an evidence tag.