agent-dashboard CLAUDE.md

agent-dashboard CLAUDE.md is an instructions file for coding agents from bjornjee/agent-dashboard. It costs 879 tokens per session, scanned A, original, MIT.

Repository-specific coding instructions for agent-dashboard, a Go project. They require external commands to be accessed through replaceable interfaces and tests to use mocks instead of starting real programs.

In plain words
What is it for?
Use them when changing command execution, writing tests, or reviewing Go code in agent-dashboard.
Why use it?
They prevent business code from being tightly tied to operating-system commands and keep tests predictable and safe.

Instructions file

Installs and runs on its own, but its text points at files inside the plugin that ships it — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed.

Install

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.

agentmods
npx agentmods add instructions/bjornjee/agent-dashboard/claude-md
Clone the repo
git clone --depth 1 https://github.com/bjornjee/agent-dashboard

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 agent-dashboard CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/bjornjee/agent-dashboard/claude-md.svg)](https://agentmods.dev/instructions/bjornjee/agent-dashboard/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/bjornjee/agent-dashboard/claude-md"><img src="https://agentmods.dev/badge/instructions/bjornjee/agent-dashboard/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 879 This file is loaded in full into every session.
When invoked 879 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.00879 $0.00879
Opus 5 $0.00439 $0.00439
Sonnet 5 $0.00176 $0.00176
Haiku 4.5 $0.00088 $0.00088

Measured 4d ago against content hash d2b135594359, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-dashboard 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 4d 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.

CLAUDE.md · 64 lines

How it starts

The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.

agent-dashboard

Go conventions

External commands must go through interfaces

Never call exec.Command or exec.CommandContext directly in business logic. All external command execution must go through a mockable interface with a package-level variable that tests can swap.

Pattern:

// 1. Define the interface in a runner.go file
type Runner interface {
    Output(ctx context.Context, name string, args ...string) ([]byte, error)
}

// 2. Production implementation
type execRunner struct{}
func (r *execRunner) Output(ctx context.Context, name string, args ...string) ([]byte, error) {
    return exec.CommandContext(ctx, name, args...).Output()
}

// 3. Package-level variable
var runner Runner = &execRunner{}

The only files allowed to import "os/exec" are runner/interface implementation files (e.g. runner.go, tmux.go). If you need a new command execution pattern, extend the existing interface rather than calling exec.Command directly.

Tests must use mocks, never real subprocesses

Tests must never spawn real subprocesses (git, gh, tmux, open, etc.). Always use mockery-generated mocks from internal/mocks/. Swap the package-level runner in tests:

m := mocks.NewMockGitRunner(t)
orig := gitRunner
gitRunner = m
t.Cleanup(func() { gitRunner = orig })

Mocks are generated by mockery — config is in .mockery.yaml. After changing an interface, regenerate with mockery.

Testing

  • Run tests with CGO_ENABLED=0 go test -race ./... (CGO disabled to avoid AMFI kills on macOS)
  • The -race flag is mandatory — it catches data races that cause tmux server crashes

Per-harness packages

The dashboard supports two coding-agent harnesses today: claude and codex. Per-harness logic lives in internal/harness/<name>/ for spawn-command construction and internal/<harness>/ for reading the harness-specific state files. Adding another harness means:

  1. Implement domain.Harness under internal/harness/<name>/ with *runner.go for any os/exec calls (same conventions as above).
  2. Register the name in internal/harness/registry.go Resolve().
  3. Add a <Name>HarnessSettings struct to internal/domain/types.go and wire it into HarnessSettings.
  4. Route per-harness flags in internal/harness/spawnopts.go so unused fields stay zero-valued.
  5. If the harness has its own conversation/state schema, add a reader under internal/<harness>/ and route through internal/conversation/router.go Read() (dispatches on agent.Harness).
  6. Update hook scripts (adapters/claude-code/scripts/hooks/agent-state-fast.js detectHarness) so the SessionStart hook stamps the harness name in the per-agent state file — without that, the router can't dispatch.

Read the full file on GitHub · 64 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. 4d ago First seen · 64 lines · 879 tokens per session scan A d2b135594359

Subscribe to this mod's changes

agent-dashboard CLAUDE.md is an instructions file published in the GitHub repository bjornjee/agent-dashboard (21 stars, last pushed 1mo ago), licensed MIT. It adds 879 tokens to every session, about $0.0044 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 2026-08-30.