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 instructions/peerjakobsen/smartspender/claude-mdgit clone --depth 1 https://github.com/peerjakobsen/smartspenderWrote 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/instructions/peerjakobsen/smartspender/claude-md)<a href="https://agentmods.dev/instructions/peerjakobsen/smartspender/claude-md"><img src="https://agentmods.dev/badge/instructions/peerjakobsen/smartspender/claude-md.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 | $0.00435 | $0.00435 |
| Opus 5 | $0.00217 | $0.00217 |
| Sonnet 5 | $0.00087 | $0.00087 |
| Haiku 4.5 | $0.00044 | $0.00044 |
Grade A, and why
smartspender CLAUDE.md 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 5d 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
SmartSpender
A Claude Cowork plugin for personal finance management. Danish bank customers. File-based architecture -- all components are markdown and JSON.
Project Structure
agent-os/-- Product spec, standards, and planning (development layer).claude-plugin/-- Plugin manifest (Cowork runtime)commands/-- Slash commands (user-invoked workflows)skills/-- Domain knowledge (auto-activated by Claude)banks/-- Bank adapter (Enable Banking API)subscriptions/-- Subscription service cancellation instructionsinvoice-knowledge/-- Learned vendor-specific invoice parserspayslip-knowledge/-- Learned employer-specific payslip parserstools/-- Helper scripts (eb-api.py for Enable Banking API). Not included in plugin zip.
Development Workflow
Author in Claude Code. Test in Cowork.
Plan → /create-spec to design the command or skill
Author → Write the markdown files following agent-os/standards/
Commit → Git commit so you can roll back
Package → zip -r smartspender.zip .claude-plugin/ commands/ skills/ banks/ subscriptions/ payslip-knowledge/
Upload → Ask the user to upload smartspender.zip as a plugin in Claude Cowork
Test → Invoke the command in Cowork, watch what happens
Log → Record what worked and what broke in TEST_LOG.md
Refine → Fix the instructions based on test results
Repeat from Commit
Build one vertical slice at a time. Get /smartspender:sync working end-to-end before starting /smartspender:analyze.
Key Constraints
- No application code -- only markdown and JSON files
- You're debugging instructions, not code. If Claude does the wrong thing, the fix is clearer markdown.
- Bank sync workflows require live MitID login -- can't be tested without the real bank site
- All data lives as local CSV files in the working directory
- Danish language for user-facing output, English for internal plugin files
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.
- 5d ago First seen · 42 lines · 435 tokens per session scan A 75570adb2a50
smartspender CLAUDE.md is an instructions file published in the GitHub repository peerjakobsen/smartspender (5 stars, last pushed 7mo ago), licensed MIT. It adds 435 tokens to every session, about $0.0022 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 instructions, from other repositories
catatz AGENTS.md
AGENTS.md instructions for kupzed/catatz, covering catatz - codex operating instructions, origin, source of truth, ecc workflow overlay and design contract.
expense-budget-tracker AGENTS.md
AGENTS.md instructions for kirill-markin/expense-budget-tracker, covering expense-budget-tracker, rules, components, supported clients and key paths.
wealth-forge-ai CLAUDE.md
Claude Code instructions for vikisingh23/wealth-forge-ai, covering wealthforge ai: claude instructions, persona, user profile, global mandates and agent specialization mapping.
wealth-forge-ai GEMINI.md
Gemini CLI instructions for vikisingh23/wealth-forge-ai, covering wealthforge ai, agent instructions, financial rules & standards and configuration.
AI-Prompt-Guide-MCP CLAUDE.md
Claude Code instructions for Blakeem/AI-Prompt-Guide-MCP, a project described as: Deprecated. MCP server and Claude Code plugin with task driven agent orchestration workflows, guides, markdown knowledge graphs, with intelligent context injection.
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.