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 agentmods add skills/cheswickdev/cairn/cairn-resumenpx skills add CheswickDEV/cairn --skill cairn-resumegit clone --depth 1 https://github.com/CheswickDEV/cairnWrote 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/cheswickdev/cairn/cairn-resume)<a href="https://agentmods.dev/skills/cheswickdev/cairn/cairn-resume"><img src="https://agentmods.dev/badge/skills/cheswickdev/cairn/cairn-resume.svg" alt="Measured on agentmods" 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 | $0.00065 | $0.00370 |
| Opus 5 | $0.00032 | $0.00185 |
| Sonnet 5 | $0.00013 | $0.00074 |
| Haiku 4.5 | $0.00006 | $0.00037 |
Grade A, and why
Cairn Resume 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 4d 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.
What it actually says
Cairn resume — continue from the ledger, don't re-read the repo
Trigger: the user says "Cairn resume", or you are picking up prior work at the start of a task.
Do this:
- Call
decision_log(view:"current") to re-inject the accepted + open decisions and the latest handoff brief. - Treat that re-injected state as the authoritative source of truth. Continue from its NEXT STEPS.
- Do NOT re-read the repo wholesale — not
ADR-cairn.md, notCHANGELOG.md, not the source tree. Open ONLY the specific files named in NEXT STEPS, or the files you are about to change. Re-reading the repo is exactly the context bloat the ledger exists to prevent — the brief is the trusted summary. - Optional: call
context_statusto see the current zone (Codex has no statusline/hooks, so check it yourself; pass the host-reportedmodel_context_window, e.g. 258400, assurfaceCap). If you need a value byte-exact that the brief references, pull its evidence withdecision_logview:"all".
When you reach a natural stopping point, or the zone turns yellow/red, run "Cairn Handoff" (skill
cairn-handoff) to persist progress.
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.
- 4d ago First seen · 24 lines · 65 tokens per session scan A c2db7669a386
Cairn Resume is a skill published in the GitHub repository CheswickDEV/cairn (1 stars, last pushed 2mo ago), licensed MIT. It adds 65 tokens to every session and 370 once invoked, about $0.0003 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.
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.
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…
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…
rlm
The RLM pattern enables processing massive contexts (10M+ tokens) that exceed Claude's context window by recursively chunking, processing, and aggregating results. Instead of failing on large files, use RLM to break them into manageable pieces.
context-compactor
Keep long Claude Code sessions coherent by rolling old conversation into a single, constantly-updated summary so the context window stops overflowing and the assistant stops giving degraded or off-topic answers. Use this skill whenever a coding or development session has been running long, whenever the conversation…
recall-maintain
Pressure-based memory cleanup. Demotes, archives, or pins entries when working memory exceeds budget.