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/0xsoftboi/suwappubot/claude-mdgit clone --depth 1 https://github.com/0xSoftBoi/suwappubotWhat 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 | $0.04935 | $0.04935 |
| Opus 5 | $0.02468 | $0.02468 |
| Sonnet 5 | $0.00987 | $0.00987 |
| Haiku 4.5 | $0.00494 | $0.00494 |
Grade A, and why
suwappubot CLAUDE.md scanned grade A with 1 finding 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 today.
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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl https://api.suwappu.bot/health # api-ts (prod) How it starts
The opening of the file, as written. The whole thing — 276 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Workflow
- IMPLEMENT, don't plan. When asked to fix or build something, DO the work. If you need to explore first, limit exploration to 5 minutes then start building. Only produce a plan document if explicitly asked for one.
- If blocked, say so explicitly — don't fill the response with exploration as a substitute for implementation.
Response Length
- Keep each assistant turn under ~400 output tokens. Prefer many short turns over one long turn.
- For long reports, audits, or plans:
Writethem to a file and reply with the path + a 3-bullet summary. Never inline a whole document in a turn. - Write deliverables to disk as they are produced, not at the end. An interrupted session should cost a turn, not the whole run.
Git Conventions
- IMPORTANT: Do NOT add "Co-Authored-By" lines to commit messages.
- Journal-only commits (
harness(journal): ...) must end the subject with[skip ci]. The test workflow has no path filters and cancels superseded PR runs, so a plain journal push cancels the in-flight CI run you're waiting on. Never put[skip ci]on a commit that touches code.
Git Operations
Before any git push, rebase, or merge, run this mandatory pre-flight sequence. Do NOT skip steps. Only proceed after reporting all findings. If any issue is found, fix it first and re-run.
Pre-flight checklist (run every time):
- Build artifacts: Check for
.next/,node_modules/,dist/in tracked files. If found, add to.gitignoreand unstage before proceeding. - Lock files: Run
ls .git/*.lock 2>/dev/null— if stale lock files exist, investigate what holds them (don't just delete). - Worktree check: Run
git rev-parse --git-common-dir— if in a worktree, NEVER rebase. Always usegit mergeorgit pull --no-rebase. - Divergence check: Compare
git rev-parse HEADvsgit rev-parse @{u}to detect local/remote divergence. Recommend merge (not force-push) unless user explicitly approves. - Uncommitted work: Run
git statusandgit stash listto surface any uncommitted changes or stashed work. Report before proceeding.
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.
- today Changed · +1 lines · +69 tokens per session 8fd85188c2c3
- yesterday First seen · 275 lines · 4,866 tokens per session scan A 5187c8e9126b
suwappubot CLAUDE.md is an instructions file published in the GitHub repository 0xSoftBoi/suwappubot (3 stars, last pushed today), licensed Apache-2.0. It adds 4,935 tokens to every session, about $0.0247 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
quantum-shield CLAUDE.md
Claude Code instructions for kota1026/quantum-shield, covering quantum shield, project structure, universal rules, domain rules and quick start.
clicks-protocol CLAUDE.md
Claude Code instructions for clicks-protocol/clicks-protocol, covering clicks protocol — agent brief, live contracts (base mainnet, owned by safe), wallets, erc-8004 trustless agent and prototypes (repo only, nicht deployed).
ponscli CLAUDE.md
Claude Code instructions for mesutgulecen/ponscli, covering ponscli, the rule that matters most, layout, verifying a change and things that will bite.
Agent-skills AGENTS.md
Instructions for EmblemCompany/Agent-skills, covering agents.md - emblemai agent skills, core emblemai message, skills directory, cross-platform skills and use-case skills.
PROGRAMMABLE AGENTS.md
AGENTS.md instructions for programmablehq/PROGRAMMABLE, covering programmable workspace rules, canonical repository, parallel work, path ownership and release discipline.
mcp AGENTS.md
AGENTS.md instructions for rustok-org/mcp, covering agents.md — rustok mcp, stack, repository rules, architecture and gates.