harness-handoff
Skill1 starsChanged 20 days agoApache-2.0View on GitHub
Use when a task is long, context is getting crowded, work is being paused, or another agent/session must continue. Produces a compact, deterministic handoff with state, decisions, files, commands, failures, and the next action.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- wyattjoh/skillstheir own version, first seen Sep 27, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- wyattjoh/skills★ 3same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository7 other files it gives its agents.
- rules/AGENTS.mdAGENTS.md
- harness-controlSkill
- harness-recallSkill
- harness-reflectorSkill
- harness-skill-authorSkill
- systematic-debuggingSkill
- verification-before-completionSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 70
No secrets: Remove credentials, tokens, cookies, and private URLs.
- Distinctive: line 79
Put durable rules in AGENTS.md, not task status.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Harness² Handoff
Long tasks fail in familiar ways: old assumptions survive, work is repeated, constraints are forgotten, and the context fills with raw output. Write a handoff before that happens.
When to write one
- The user wants to pause, change machines, or continue later.
- The task will span more than one session or agent.
- Most of the context is exploration rather than decisions.
- A debugging attempt is unfinished and must be resumed.
Template
# Handoff: <task>
## Goal
One paragraph, including the user's acceptance criteria.
## Current state
- Done:
- In progress:
- Not started:
- Verified by:
## Decisions
- Decision — reason — alternatives rejected.
## Files
- `path` — what changed and why.
- `path` — context only, no changes.
## Commands
- `<command>` — passes or fails, with a short output summary.
## Failures and hypotheses
- Error — evidence — ruled out or still suspected.
## Constraints
- Rules from AGENTS.md, user preferences, compatibility requirements.
## Open questions
- Question — who can answer it, or which experiment will settle it.
## Next action
The single next concrete step, with the exact command or file.
## Verification
How the next session proves the task is complete.Rules
- Facts over story. Include exact commands, error signatures, paths, and versions. Summarize or link raw logs.
- Keep proof and guesses apart. A hypothesis must not read like a confirmed cause.
- Carry the constraints forward. Copy the relevant
AGENTS.mddirectives and user requirements into the handoff. - No secrets. Remove credentials, tokens, cookies, and private URLs.
- One next action. Five competing next steps is not a handoff.
- Stay small. Aim for one page. Move reference material into
docs/and link it.
After writing
- Save it in the repository, for example
docs/HANDOFF.md, if the project will use it. Otherwise return it in the conversation for a short pause. - Put durable rules in
AGENTS.md, not task status. - Record lasting verified lessons with Harness² and keep temporary state in the handoff.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

