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.
npx skills add 44-pixels/handover-mcp --skill handoffgit clone --depth 1 https://github.com/44-pixels/handover-mcpWrote 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.
[](https://agentmods.dev/skills/44-pixels/handover-mcp/handoff)<a href="https://agentmods.dev/skills/44-pixels/handover-mcp/handoff"><img src="https://agentmods.dev/badge/skills/44-pixels/handover-mcp/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.
<a href="https://agentmods.dev/skills/44-pixels/handover-mcp/handoff"><img src="https://agentmods.dev/badge/skills/44-pixels/handover-mcp/handoff.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00065 | $0.01103 |
| Opus 5.5 | $0.00026 | $0.00441 |
| Sonnet 5.5 | $0.00013 | $0.00221 |
| Haiku 4.5 | $0.00006 | $0.00110 |
Grade A, and why
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 2mo ago.
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.
How it starts
The opening of the file, as written. The whole thing — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hand off work that another actor can continue
Create the smallest complete continuation record. The next human or agent should be able to identify the goal, trust the evidence, and take one useful action without receiving the original chat.
This skill works locally. Handover MCP or the handover-sh CLI adds shared,
versioned publishing when it is available; never make installation a
prerequisite for creating a useful handoff.
Choose create or resume
- Create when the current actor is stopping, changing models, approaching a context limit, or transferring ownership.
- Resume when the user provides
HANDOFF.md, a Handover link, a handover slug, or another continuation record. - If the request needs strict machine-readable JSON, schema validation, or
protocol exchange, use the companion
handover-recordskill.
Inspect the source of truth
Before writing, inspect the current task, relevant files, version-control status and diff, linked issue or pull request, test output, and any existing handoff. Prefer original artifacts over remembered summaries. Update an existing canonical handoff instead of creating a duplicate.
Do not invent missing state. Label it Unverified or make it an open question.
Keep credentials, secrets, tokens, private keys, copied sessions, hidden
reasoning, and unnecessary personal data out of every artifact.
Create the local record
Use the path requested by the user. Otherwise update an existing handoff or
write HANDOFF.md at the project root when the record should travel with the
project.
# Handoff
## Goal
The intended outcome and acceptance criteria.
## Current state
- Complete:
- In progress:
- Blocked:
- Unverified:
## Decisions
- Decision, reason, and consequence.
## Evidence
- File, URL, command result, test, revision, or other original artifact.
## Next action
The smallest useful next step, with an owner when known.
## Open risks and questions
- Anything that could change the next action.
## Relevant artifacts
- Exact paths or links the next actor should inspect.
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.
- 2mo ago First seen · 133 lines · 65 tokens per session scan A cc215a96e0d4
handoff is a skill published in the GitHub repository 44-pixels/handover-mcp (3 stars, last pushed 3mo ago), licensed MIT. It adds 65 tokens to every session and 1,103 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-31.
handoff compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| handoff44-pixels/handover-mcp | Skill | 3 | 3mo ago | MIT | A | 65 | Claude Code, Codex |
| code-reviewShy2593666979/AgentChat | Skill | 930 | 21d ago | MIT | A | 41 | Claude Code, Codex |
| data-analysisShy2593666979/AgentChat | Skill | 930 | 21d ago | MIT | A | 34 | Claude Code, Codex |
| research-reportShy2593666979/AgentChat | Skill | 930 | 21d ago | MIT | A | 35 | Claude Code, Codex |
| email-writingShy2593666979/AgentChat | Skill | 930 | 21d ago | MIT | A | 37 | Claude Code, Codex |
| knowledge-organizingShy2593666979/AgentChat | Skill | 930 | 21d ago | MIT | A | 41 | Claude Code, Codex |
| product-documentShy2593666979/AgentChat | Skill | 930 | 21d ago | MIT | A | 33 | Claude Code, Codex |