Cairn Handoff

Cairn Handoff is a skill for Claude Code, Codex from CheswickDEV/cairn. It costs 86 tokens per session (501 once invoked), scanned A, original, MIT.

A session handoff record organised into seven sections, including decisions, evidence, open questions, constraints, and next steps. Cairn is the system used to save that record so work can continue in another session or environment.

In plain words
What is it for?
Use it before clearing or switching a session, or when work needs to be passed to another agent or person. It records unresolved work, exact values, discarded options, and the actions still needed.
Why use it?
It prevents important context from being lost when a session ends, changes model, or reaches its context limit. Evidence references make it possible to check where key facts came from.

Skill for Claude CodeCodex

Part of the cairn plugin — 4 skills, 2 hooks, 1 MCP server shipped together

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.

agentmods
npx agentmods add skills/cheswickdev/cairn/cairn-handoff
Any agent
npx skills add CheswickDEV/cairn --skill cairn-handoff
Clone the repo
git clone --depth 1 https://github.com/CheswickDEV/cairn

Made for: Claude Code, Codex.

Or install cairn, the plugin that ships this one along with the rest of its 4 skills, 2 hooks, 1 MCP server.

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 Cairn Handoff

README.md
[![agentmods](https://agentmods.dev/badge/skills/cheswickdev/cairn/cairn-handoff.svg)](https://agentmods.dev/skills/cheswickdev/cairn/cairn-handoff)
Your own site
<a href="https://agentmods.dev/skills/cheswickdev/cairn/cairn-handoff"><img src="https://agentmods.dev/badge/skills/cheswickdev/cairn/cairn-handoff.svg" alt="Measured on agentmods" height="20"></a>
Per session 86 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 501 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00086 $0.00501
Opus 5 $0.00043 $0.00251
Sonnet 5 $0.00017 $0.00100
Haiku 4.5 $0.00009 $0.00050

Measured 3d ago against content hash dc1fa76b1fba, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Cairn 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 3d 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.

.agents/skills/cairn-handoff/SKILL.md · 30 lines

What it actually says

Cairn Handoff — persist this session as a 7-bucket brief

Trigger: the user says "Cairn Handoff", or you are about to clear/finish the session or switch model/host.

Do this:

  1. Write a faithful 7-bucket brief from THIS conversation — nothing invented, nothing padded:
    • DECISIONS (mark superseded ones), EVIDENCE (each with a source ref: file:line, msg#, url, or tool output), OPEN QUESTIONS, CONSTRAINTS, VERBATIM (code/IDs/values/paths byte-exact), NEXT STEPS, DISCARDED.
  2. Call the handoff MCP tool (leave mode at its default account → no model call, no egress; on Codex you write the brief on the user's ChatGPT/Codex subscription):
    • brief: the full brief text.
    • requiredVerbatim: every value that MUST survive byte-exact (IDs, exact numbers, file paths, flags).
    • decisions / evidence: the structured form when you can extract it.
    • host: the current model/provider; sourceZone: the current zone (green/yellow/red).
  3. Report back: storedDecisionId, decisionsStored, and missingRequired. If missingRequired is non-empty, fix the brief so those spans appear byte-exact and call handoff again.

Notes:

  • A new handoff supersedes the previous brief automatically, so the ledger stays lean — old briefs do not pile up in decision_log view=current.
  • Account mode persists locally only; nothing is sent to any model.
  • To pick the work back up later, use "Cairn resume" (skill cairn-resume).
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. 3d ago First seen · 30 lines · 86 tokens per session scan A dc1fa76b1fba

Subscribe to this mod's changes

Cairn Handoff is a skill published in the GitHub repository CheswickDEV/cairn (1 stars, last pushed 2mo ago), licensed MIT. It adds 86 tokens to every session and 501 once invoked, about $0.0004 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 2026-08-31.

Related

Other skills, from other repositories

mnemo-cortex

Installs and wires Mnemo Cortex (local-first persistent memory) into OpenClaw and other MCP-capable agents. Use for cross-session recall, decision history, or multi-agent shared memory.

GuyMannDude/mnemo-cortex · 44 tokens

vibebox

Use this skill when an AI coding task should consult VibeBox memory before work as an auto-intervention memory layer for meaningful repository work; retrieve active guidance when available, apply it as context, and capture meaningful outcomes with userRequest plus AI-agent structured memory candidates unless the user…

boksajang/vibebox · 62 tokens

coss-particles

Index of all COSS UI particle examples. Use when implementing UI features to find copy-paste-ready component patterns built on coss primitives. Each particle has a description and a JSON URL for easy installation.

CherryHQ/stella · 46 tokens

skill-creator

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

CherryHQ/stella · 64 tokens

lark-cli

Lark/Feishu CLI skills for Stella sessions. Covers workspace operations — calendar, docs, tasks, mail, and messenger — via the Stella-managed lark-cli tool. Stella installs the binary and injects a managed per-user OAuth token; no native CLI bootstrap is required. Always read lark-shared first for identity selection…

CherryHQ/stella · 79 tokens

stella

Self-knowledge about stella, the self-hosted AI assistant. Use when the user asks about stella itself: configuration, setup, onboarding, providers, models, agents, channels (Telegram/Discord/QQ/Feishu/DingTalk/WeChat), webhooks, memory system (LCM), scheduled jobs, reusable workflows, goals (objectives that converge…

CherryHQ/stella · 310 tokens