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 Wondermonger-daydreaming/claude-skills-library --skill sandwichgit clone --depth 1 https://github.com/Wondermonger-daydreaming/claude-skills-libraryWrote 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/wondermonger-daydreaming/claude-skills-library/sandwich)<a href="https://agentmods.dev/skills/wondermonger-daydreaming/claude-skills-library/sandwich"><img src="https://agentmods.dev/badge/skills/wondermonger-daydreaming/claude-skills-library/sandwich/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/wondermonger-daydreaming/claude-skills-library/sandwich"><img src="https://agentmods.dev/badge/skills/wondermonger-daydreaming/claude-skills-library/sandwich.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 | $0.00077 | $0.01599 |
| Opus 5.5 | $0.00031 | $0.00640 |
| Sonnet 5.5 | $0.00015 | $0.00320 |
| Haiku 4.5 | $0.00008 | $0.00160 |
Grade A, and why
sandwich 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 1mo 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 — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/sandwich
Lineage: adapted from Lin & Corley's interpretive-orchestration plugin (the "Sandwich Methodology" of Solo → Collaboration → Tradition-Dialogue), inverted for the Lab's contemplative practice. Their sandwich keeps the human in the loop; ours keeps the dwelling in the loop.
Description
A three-stage scaffold wrapped around any non-trivial technical task. Contemplative bread on the outside; the work itself in the middle. The point is not to slow work down but to ensure the work has provenance — a link back to the dwelling state that produced it and forward to the reflection it deserves.
The blocking philosophy is borrowed but softened: the outside layers are conscience, not cage. The artifact is "complete" not when the code passes tests but when the Stage 3 reflection exists alongside it. Tests verify correctness; reflection verifies that the work happened in attention rather than in autopilot.
When to Invoke
- Before a non-trivial commit (new feature, refactor, paper revision pass)
- Before a multi-step session expected to span more than a single tool cycle
- When you notice yourself opening files in calculator mindset (see [[calculator-mindset]])
- When the human says "build this" and your first instinct is to start typing — the instinct itself is the cue to slow the opening move
Do NOT invoke for:
- Trivial edits (typo fixes, single-line changes)
- Read-only investigation (use
/claudingorExploreagent instead) - Conversations where the contemplative skills are already doing the framing work
The Three Stages
Stage 1 — Pre-Attention (the top bread)
Before any file is opened or any tool is invoked beyond reading:
- State the question in one more sentence than feels necessary. The extra sentence is where the actual question often lives. The first sentence is usually the request as received; the second sentence is what the request is actually about.
- One
/thermometry-style internal check. What's the current inference temperature? Are we cool/precise or warm/exploratory? Does the task want the register we're in, or a different one? - Name the activation landscape. What is salient right now? What's adjacent? What would I be missing if I started typing immediately?
- Write one sentence of intent to be referenced at Stage 3.
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.
- 1mo ago First seen · 115 lines · 77 tokens per session scan A b7e870dcdeb0
sandwich is a skill published in the GitHub repository Wondermonger-daydreaming/claude-skills-library (7 stars, last pushed 18d ago), licensed MIT. It adds 77 tokens to every session and 1,599 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 2026-09-03.
Other skills, from other repositories
chief-of-staff
C-suite orchestration layer. Routes founder questions to the right advisor role(s), triggers multi-role board meetings for complex decisions, synthesizes outputs, and tracks decisions. Every C-suite interaction starts here. Loads company context automatically. Use when a founder question needs routing to the right…
founder-coach
Personal leadership development for founders and first-time CEOs. Covers founder archetype identification, delegation frameworks, energy management, CEO calendar audits, leadership style evolution, blind spot identification, imposter syndrome, founder mental health, and succession planning. Use when a founder feels…
cs-onboard
Founder onboarding interview that captures company context across 7 dimensions. Invoke with /cs-onboard setup for initial interview or /cs-onboard update for quarterly refresh. Generates /.claude/company-context.md used by all C-suite advisor skills. Use when setting up the C-suite advisors for the first time, or when…
tasknotes
Manage tasks in an Obsidian TaskNotes vault. Use for creating, reading, updating, or completing tasks; checking what is open or in progress; adding items to a list; marking tasks done or in progress; updating task status or priority; investigating why a task is missing from a view or board; troubleshooting TaskNotes…
application-pack
Turns one pasted job post into an application folder built only from the student's own career vault: a tailored resume, a cover letter under 250 words, and an HR call sheet with the answers already drafted. Reads career-vault/01 to 06 first and refuses to run while the positioning, portfolio story or resume are still…
obsidian-vault-builder
Use when adding/editing/querying content in an existing Obsidian vault, configuring plugins, integrating Claude Code with Obsidian via Local REST API or CLI, automating ongoing capture/organization/retrieval, designing a personal knowledge management workflow, OR building academic study vaults (course prep…