costguard-audit

costguard-audit is a command for Claude Code from mbanderas/costguard. It costs 34 tokens per session (499 once invoked), scanned A, original, MIT.

A command that checks registered workspaces for wasted CI minutes and scheduled jobs, with optional read-only checks of cloud-provider billing. CI means automated jobs that test or build code; cron jobs are scheduled tasks.

In plain words
What is it for?
Auditing one workspace or all workspaces, narrowing checks to CI or scheduled tasks, checking supported provider billing when credentials are available, and producing Markdown or JSON reports.
Why use it?
It identifies avoidable usage and estimates the monthly cost of each finding, so teams can focus on the largest sources of waste.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the costguard plugin — 1 skill, 4 commands, 1 MCP server shipped together

Good fit Auditing one workspace or all workspaces, narrowing checks to CI or scheduled…

Compare 6 commands from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add mbanderas/costguard
Claude Code
/plugin install costguard

Made for: Claude Code.

Or install costguard, the plugin that ships this one along with the rest of its 1 skill, 4 commands, 1 MCP server.

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 costguard-audit

README.md
[![agentmods](https://agentmods.dev/badge/commands/mbanderas/costguard/costguard-audit.svg)](https://agentmods.dev/commands/mbanderas/costguard/costguard-audit)
Your own site
<a href="https://agentmods.dev/commands/mbanderas/costguard/costguard-audit"><img src="https://agentmods.dev/badge/commands/mbanderas/costguard/costguard-audit.svg" alt="Measured on agentmods" height="20"></a>
Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 499 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00034 $0.00499
Opus 5 $0.00017 $0.00249
Sonnet 5 $0.00007 $0.00100
Haiku 4.5 $0.00003 $0.00050

Measured 6d ago against content hash bd50d66e8d94, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

costguard-audit 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 6d 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.

commands/costguard-audit.md · 43 lines

What it actually says

Run a Costguard cost audit and report the findings.

Requested action: $ARGUMENTS

Costguard is a built CLI bundled with this plugin at ${CLAUDE_PLUGIN_ROOT}/dist/cli/index.js. It reads a workspaces.json registry from the current working directory — run from a repo that has one (e.g. the costguard checkout, or any project where you ran costguard registry init).

  1. Run the audit, passing the user's arguments straight through:

    node "${CLAUDE_PLUGIN_ROOT}/dist/cli/index.js" audit $ARGUMENTS
    
    • No workspace named and no --all: the CLI lists what is available — pass a workspace name (e.g. web-app) or --all.
    • --providers all (or a comma list like --providers github,netlify) adds read-only cloud billing checks; provider modules whose token env var is absent are skipped, not failed.
    • --ci-only / --crons-only narrow the static checks; --json emits JSON.
  2. Report the CLI's stdout verbatim — it is the cost report (estimated monthly waste per finding, with a fix suggestion). On a non-zero exit, relay the Error: <message> line.

Notes:

  • Prerequisite: the plugin runs the built dist/. If ${CLAUDE_PLUGIN_ROOT}/dist/cli/index.js is missing, build it once from the costguard checkout with pnpm build (then pnpm install if deps are absent).
  • All billing checks are read-only (GET / read-only GraphQL only). Costguard never writes to provider accounts and never prints tokens.
  • Tokens are read from the environment / .env only. A workspace's providers are listed in workspaces.json; a provider with no token present is skipped.
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. 6d ago First seen · 43 lines · 34 tokens per session scan A bd50d66e8d94

Subscribe to this mod's changes

costguard-audit is a command published in the GitHub repository mbanderas/costguard (2 stars, last pushed 3d ago), licensed MIT. It adds 34 tokens to every session and 499 once invoked, about $0.0002 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.