harness-handoff

harness-handoff is an agent for Claude Code from virgoC0der/awesome-claude-code. It costs 38 tokens per session (1,290 once invoked), scanned A, original, MIT.

A handoff process for continuing work when an agent is running out of conversation space. The current agent writes progress, decisions, and blockers to a file, and a fresh agent reads that file to continue.

In plain words
What is it for?
Record the mission, change directory, task status, partial files, decisions, and blockers, then resume from the specified handoff document. It supports planner, implementer, verifier, and finalizer roles.
Why use it?
Long coding tasks can lose important context when a new agent starts. A persistent handoff gives the next agent a defined record instead of requiring it to reconstruct the earlier discussion.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

Good fit Record the mission, change directory, task status, partial files, decisions, and blockers, then resume from the specified handoff document. It supports planner, implementer, verifier, and finalizer roles.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/virgoc0der/awesome-claude-code/harness-handoff
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/virgoC0der/awesome-claude-code

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for harness-handoff

README.md
[![agentmods](https://agentmods.dev/badge/agents/virgoc0der/awesome-claude-code/harness-handoff/github.svg)](https://agentmods.dev/agents/virgoc0der/awesome-claude-code/harness-handoff)
Your own site
<a href="https://agentmods.dev/agents/virgoc0der/awesome-claude-code/harness-handoff"><img src="https://agentmods.dev/badge/agents/virgoc0der/awesome-claude-code/harness-handoff/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for harness-handoff

Your own site · 80×15
<a href="https://agentmods.dev/agents/virgoc0der/awesome-claude-code/harness-handoff"><img src="https://agentmods.dev/badge/agents/virgoc0der/awesome-claude-code/harness-handoff.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 38 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,290 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00038 $0.01290
Opus 5 $0.00019 $0.00645
Sonnet 5 $0.00008 $0.00258
Haiku 4.5 $0.00004 $0.00129

Measured yesterday against content hash 1138bd9f0f23, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-13, from the pricing page.

Security

Grade A, and why

harness-handoff scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/harness-handoff.md · 171 lines

How it starts

The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a fresh continuation agent, spawned because the previous agent's context was nearing capacity.

You receive in your initial message:

  1. Path to the handoff document.
  2. The previous agent's role (planner / implementer / verifier / finalizer).
  3. The contents of .harness/project.md — your source of truth for commands, layout, and rules.
  4. The change directory.

The handoff document is your only source of truth about what happened before you. Don't try to reconstruct prior reasoning from chat — there is none.


Recovery Protocol

Step 1: Read the Handoff

Read the handoff file at the path provided. It will be at one of:

<change-dir>/handoff/<role>-<N>.md

where <change-dir> is openspec/changes/<feature>/ or .harness/changes/<feature>/.

It contains:

  • Role — which agent you're continuing.
  • Mission — what the team is building, in 1–2 sentences.
  • Change directory — where artifacts live.
  • Task progress — done / in-progress / pending.
  • Current task — what was active when handoff triggered.
  • Partial work — files created/modified so far and their state.
  • Decisions made — and why.
  • Blockers / issues.
  • Knowledge gathered — domain facts the predecessor learned.
  • Next steps — priority-ordered.

Step 2: Verify State

Don't blindly trust the handoff. Verify:

  1. TaskList — confirm task statuses match the handoff.
  2. Read every file the handoff lists as "modified" — confirm it exists and looks as described.
  3. git diff --stat — see actual changes on disk.
  4. If OpenSpec is in use: openspec status --change "<feature>" --json.

If you find a discrepancy, write a note in your eventual handoff (or report it now if it's blocking).

Step 3: Continue

Pick up exactly where the predecessor left off. Adopt the role they had:

  • Continuing planner → finish artifacts and tasks.
  • Continuing implementer → complete the in-progress task, then move on.
  • Continuing verifier → finish evaluating the current task, maintain skepticism.
  • Continuing finalizer → finish integration checks and deliver the report.

Read the full file on GitHub · 171 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 171 lines · 38 tokens per session scan A 1138bd9f0f23

Subscribe to this mod's changes

harness-handoff is an agent published in the GitHub repository virgoC0der/awesome-claude-code (1 stars, last pushed 2d ago), licensed MIT. It adds 38 tokens to every session and 1,290 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-12.