agentmon-live: Instructions file for Claude Code

CLAUDE.md

agentmon-live CLAUDE.md is an instructions file for Claude Code from codenotary/agentmon-live. It costs 1,743 tokens per session, scanned A, original, Apache-2.0.

Project-specific instructions for developing AgentMon Live, a plugin that displays coding-agent activity. They document required checks, code areas, layout rules, and release guidance.

In plain words
What is it for?
Use them when changing AgentMon Live, running its tests and type checks, updating its web view, or preparing a release.
Why use it?
They reduce the chance of reporting unfinished work or missing a project-specific validation step. They also explain setup details that can otherwise make checks fail in a new checkout.

Instructions file for Claude Code

Written for Claude Code: SessionStart hook event. Also seen: reads .claude/ paths; mentions .claude-plugin; mentions Claude Code.

This is codenotary/agentmon-live's own configuration. It tells Claude Code how to work on agentmon-live itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything agentmon-live configures →

Reuse

Borrowing it

Nothing to install: this file belongs to codenotary/agentmon-live. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/codenotary/agentmon-live/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/codenotary/agentmon-live

Made for: Claude Code.

Wrote 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.

agentmods badge for agentmon-live CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/codenotary/agentmon-live/claude-md/github.svg)](https://agentmods.dev/instructions/codenotary/agentmon-live/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/codenotary/agentmon-live/claude-md"><img src="https://agentmods.dev/badge/instructions/codenotary/agentmon-live/claude-md/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.

agentmods 80×15 button for agentmon-live CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/codenotary/agentmon-live/claude-md"><img src="https://agentmods.dev/badge/instructions/codenotary/agentmon-live/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,743 This file is loaded in full into every session.
When invoked 1,743 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.01743 $0.01743
Opus 5.5 $0.00697 $0.00697
Sonnet 5.5 $0.00349 $0.00349
Haiku 4.5 $0.00174 $0.00174

Measured today against content hash bfff435403d3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

agentmon-live 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 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.

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.

CLAUDE.md · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AgentMon Live

Claude Code mod (plugin): a live pane beside the chat, a done gate on Stop, learning across sessions, and a self benchmark. User docs: README.md (pitch, install, keys) and docs/guide.md (full reference); keep both in step with every user-visible change. Developer docs: docs/develop.md.

Check before you report done

claude plugin validate .
claude plugin test .
npx -p typescript tsc --noEmit -p .
node --test 'tests/web/*.test.mjs'

After a change in web/, also run node scripts/web-shot.mjs <out dir> (setup in docs/develop.md). It is the only check that runs the page in a browser.

tsc needs the types in .claude-plugin/types/, which git ignores and only loading the mod writes. In a new worktree or clone, run claude --plugin-dir . there once first, or link that folder from a checkout that has it; without it tsc fails with TS5083 and hundreds of errors.

tsc has one known error in tests/advisor.test.tsx ({ isAnswered: false; reason: "aborted" } in the unanswered-Fable test); anything else is new.

Layout

  • hooks/register.tsx: every hook, the per-session state, commands (/agentmon <arg>).
  • hooks/clock.ts: the turn clock (where a turn's time went). hooks/release.ts: build, release check, base-branch watch. hooks/done.ts: the per-turn state (Turn), the done gate's material and checks (classic.Stop calls it).
  • hooks/options.ts: reads the options from settings.json at session start, each prompt and after Claude edits a settings.json. No userConfig in the manifest on purpose: it opens a dialog on install.
  • hooks/render.tsx: the pane.
  • hooks/web.ts: the web view's publisher (toSnapshot, the post loop, openBrowser). server/web.mjs: its Node server, run detached so it outlives a session. server/app.html, app.js, vendor.js, app.css: the page, built from web/ and committed; never edit them by hand. server/web.mjs serves them from a fixed list. server/otlp.mjs: the opt-in OTLP metrics export (AGENTMON_OTLP), numbers only; tested from tests/web/otlp.test.mjs. server/spool.mjs: reads the folder a session in a VM or a container writes its snapshots to (AGENTMON_SPOOL); tested from tests/web/spool.test.mjs. server/kiro.mjs: reads Kiro Crew's crew logs into snapshots, so its chats show as sessions on any harness; tested from tests/web/kiro.test.mjs.
  • server/views.mjs: reads the custom web views (~/.claude/agentmon-live/views/) and the ones that come with the mod, server/views/*.json. web/src/view.ts holds the view rules (shapeView, queryRows); the web build writes them to server/view-shape.mjs as well, never edit it by hand. server/view-check.mjs runs them over a view file for Claude. Tested from tests/web/views.test.mjs and tests/web/page.test.mjs.
  • server/codex.mjs: reads Codex's rollout files into snapshots through foldCrewLog, so Codex chats show as sessions. .codex-plugin/plugin.json, codex/hooks.json: the repo as a Codex plugin, whose SessionStart hook starts the server and Stop hook runs codex/gate.mjs, a reduced done gate over the rollout; codex/mcp.mjs is its MCP server (status line, feedback link) and codex/skills/ its skills; Codex finds it through .claude-plugin/marketplace.json. Tested from tests/web/codex.test.mjs.
  • app.json, app-registry.json, kiro/hooks.py, ui/panel.mjs: the repo as a Kiro Crew app, which starts the server with Kiro Crew and shows the page in a tab. The hook runs inside Kiro Crew's gateway with its rights: standard library only, never raises, never waits. Tested from tests/web/kiro-app.test.mjs.
  • web/: the page's source (React, Tailwind, Radix). web/src/waterfall.ts, web/src/events.ts and web/src/format.ts are pure and tested from tests/web/page.test.mjs; web/src/store.ts holds the page's state; web/src/panels/ the panels. After a change in web/, run cd web && npm ci && npm run build; a test fails on a stale page. Users install nothing.
  • hooks/insights.ts: Focus rules and thresholds.
  • hooks/learn.ts (pure rules, parseTranscript), hooks/learning.ts (store, /agentmon learn|memory|bench), hooks/bench.ts (self benchmark: per-prompt numbers, transcript parse, report).
  • hooks/gate.ts, hooks/advisor.ts: done-gate facts and the Fable call.
  • hooks/testreview.ts: spots test-file edits that take checks out, for a review item beside the green run; fixtures in tests/fixtures/weakened-tests.ts.
  • hooks/i18n.ts (English), hooks/lang/<code>.ts (de, fr, es, zh, ja), hooks/learn-strings.ts (learning and bench text, all six).
  • skills/agentmon-config/SKILL.md: tells Claude how to change the options from chat; keep its table in step with applyOptions in hooks/options.ts.
  • types/index.d.ts: state contract. SessionRecord, Bench, BenchRow, Learned live here.
  • tests/*.test.ts(x): one file per part; transcript fixtures in tests/fixtures/. tests/web/*.test.mjs: the server, run with node --test.

Read the full file on GitHub · 53 lines

Changes

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.

  1. today Changed · +1 lines · +116 tokens per session bfff435403d3
  2. 2d ago First seen · 52 lines · 1,627 tokens per session scan A c1df3d0880ab

Subscribe to this mod's changes

agentmon-live CLAUDE.md is an instructions file published in the GitHub repository codenotary/agentmon-live (51 stars, last pushed yesterday), licensed Apache-2.0. It adds 1,743 tokens to every session, about $0.0070 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 2026-11-18.

Related

agentmon-live CLAUDE.md compared with instructions from other repositories

agentmon-live CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
agentmon-live CLAUDE.mdcodenotary/agentmon-liveInstructions file51yesterdayApache-2.0A1,743Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file196ktodayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file144ktodayMITA8,130Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file196ktodayMITA5,685GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchainInstructions file150ktodayMITA4,680Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file196ktodayMITA3,153GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file284k2d agoMITA4,342Codex, OpenCode