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/coaden/agent-kudos/agents-mdgit clone --depth 1 https://github.com/Coaden/agent-kudosWrote 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/coaden/agent-kudos/agents-md)<a href="https://agentmods.dev/instructions/coaden/agent-kudos/agents-md"><img src="https://agentmods.dev/badge/instructions/coaden/agent-kudos/agents-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 | $0.00515 | $0.00515 |
| Opus 5 | $0.00258 | $0.00258 |
| Sonnet 5 | $0.00103 | $0.00103 |
| Haiku 4.5 | $0.00052 | $0.00052 |
Grade A, and why
agent-kudos AGENTS.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 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.
How it starts
The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This repository is a public TypeScript package. Changes should be reviewable, portable, secure by default, and appropriate for an OSS audience.
Working agreement
- Read
docs/plan.md,ARCHITECTURE.md, andSECURITY.mdbefore architectural changes. - Use Node.js 22.13 or newer, ESM, strict TypeScript, and the built-in
node:sqliteAPI. - Keep V1 single-machine and local-first. Do not add hosted services, telemetry, accounts, network listeners, or postinstall behavior.
- Do not publish packages, create releases, alter npm configuration, or write into a user’s real
~/.agentsdirectory without explicit authorization. - Tests and demos must use isolated temporary homes.
- Never commit real kudos data, SQLite databases, environment files, secrets, generated coverage, or npm tarballs.
Architectural invariants
- SQLite event rows are canonical and append-only. State changes are new events.
- Agent IDs are validated stable identities; aliases never silently merge identities.
- MCP writes use the actor bound at server startup. Tool arguments cannot override it.
WINS.md, inbox files, andprofile.jsonare generated projections.NOTES.mdis human-owned and must never be overwritten.- Machine reads stay context-bounded: compact list/change feeds have conservative defaults, hard limits, byte budgets, and opaque cursors.
WINS.mdis not a machine query source. - Projection cleanup may remove only regular files listed in the generated-files manifest.
- Normal mutations must not rebuild or rewrite every pending inbox entry; full regeneration belongs to the explicit rebuild operation.
systemactors are automation identities and have no implicit agent or administrative authority.- Raw JSON/JSONL export must remain available when canonical rows are unsupported or malformed; writes must fail closed across unknown event semantics.
- Do not follow symlinks outside the configured home.
- Evidence is descriptive metadata, never captured tool output or secret material.
- No filesystem work occurs at module import time, and library code never calls
process.exit().
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.
- 3d ago First seen · 43 lines · 515 tokens per session scan A f430f958d0de
agent-kudos AGENTS.md is an instructions file published in the GitHub repository Coaden/agent-kudos (0 stars, last pushed 3d ago), licensed MIT. It adds 515 tokens to every session, about $0.0026 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
ZhiWei CLAUDE.md
Instructions for ntygod/ZhiWei, covering claude.md, project overview, build & development commands, backend and frontend.
Vault-Agent-Memory AGENTS.md
Instructions for zycaskevin/Vault-Agent-Memory, covering agent instructions for vault agent memory, common install architecture, second decision: optional features, first decision: database scope and safe agent workflow.
ZhiWei AGENTS.md
Instructions for ntygod/ZhiWei, covering repository guidelines, project structure & module organization, agent harness, build, test, and development commands and coding style & naming conventions.
claude-memory-mcp AGENTS.md
Instructions for WhenMoon-afk/claude-memory-mcp, covering mooncite repository contract, product boundary, fixed architecture, sources of truth and change discipline.
quicken-mac-mcp CLAUDE.md
Claude Code instructions for dweekly/quicken-mac-mcp, covering claude.md — developer guidelines, build & run commands and code guidelines.
grounded-knowledge-engine AGENTS.md
Instructions for dimosgit/grounded-knowledge-engine, covering agent guide, agent operating contract, authorship and attribution, two npm trees and commands — engine (repo root).