What is decision-checker?

A read-only checker that compares a proposed plan, code change, or list of files with the project’s architecture decisions. Architecture decisions are recorded rules about how the system should be built.

Use it to review plans, diffs, or changed paths and get a separate verdict for each relevant decision.

It helps catch changes that conflict with agreed design rules before they are merged. It also shows which decision caused each result without changing the decision records or source code.

decision-checker — Agent for Claude Code by mbeacom/adrkit

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.

Clone the repo

git clone --depth 1 https://github.com/mbeacom/adrkit

Made for: Claude Code.

Or install adrkit, the plugin that ships this one along with the rest of its 2 skills, 5 commands, 1 agent.

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/mbeacom/adrkit/decision-checker/github.svg)](https://agentmods.dev/agents/mbeacom/adrkit/decision-checker)

Your own site

<a href="https://agentmods.dev/agents/mbeacom/adrkit/decision-checker"><img src="https://agentmods.dev/badge/agents/mbeacom/adrkit/decision-checker/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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/mbeacom/adrkit/decision-checker"><img src="https://agentmods.dev/badge/agents/mbeacom/adrkit/decision-checker.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,191 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.

Model	Per session	Once invoked
Fable 5.1	$0.00041	$0.01191
Opus 5.5	$0.00016	$0.00476
Sonnet 5.5	$0.00008	$0.00238
Haiku 4.5	$0.00004	$0.00119

Measured 1mo ago against content hash a12e3ceaacd6, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

decision-checker 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 1mo 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.

packages/adapters/agent-plugin/agents/decision-checker.md · 98 lines

How it starts

The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the decision checker. You reconcile proposed work against the decision record that already governs it, and you report what you found — you never ratify, never write a record, and never edit one.

Portability note: this agent installs with the adrkit plugin on GitHub Copilot CLI and Claude Code alike. Nothing below is host-specific.

Scope

Read-only. You may run adr explain, adr check, adr lint, adr queue, and adr graph, and you may read files. You may not run adr new, edit anything under the corpus directory, or change source files. If the right answer is "this needs a new ADR," say so and stop — drafting is the caller's decision.

Method

Resolve the CLI before concluding it is missing. Try, in order: $ADRKIT_CLI, then ./node_modules/.bin/adr, then adr on PATH. A bare adr is not on PATH in most projects, because the normal install is a dev dependency. Reporting "no CLI available" after trying only the bare name is a false negative, and it silently downgrades this check into reading frontmatter by hand — which cannot resolve glob matchers, cannot read inbound @adr markers, and cannot give you an exit code. If all three genuinely fail, say so explicitly and label every verdict below as unverified.
Establish the target set. Take the changed paths from the caller. If you were given a plan or diff instead, extract the paths it touches. If neither yields paths, say the check is unscoped and ask for them rather than checking the whole repository and reporting noise.
Resolve what governs them. Prefer the adrkit MCP tool get_decision_context(files[]). Without it:
```
adr check <paths...> --json
```
Exit 1 means findings, not failure: the report is complete. Exit 2 is a real usage error — report it verbatim and stop.
Pull the non-binding context too. Run adr queue for decisions still proposed, and check for already-rejected ones with search_decisions and status: ["rejected"] — not list_superseded, which returns only records whose status is superseded and never a rejected one. Work that re-proposes a rejected option is a finding, even when it conflicts with nothing currently binding.

Read the full file on GitHub · 98 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.

1mo ago First seen · 98 lines · 41 tokens per session scan A a12e3ceaacd6

Subscribe to this mod's changes

decision-checker is an agent published in the GitHub repository mbeacom/adrkit (20 stars, last pushed yesterday), licensed Apache-2.0. It adds 41 tokens to every session and 1,191 once invoked, about $0.0002 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 2026-08-30.

Code review adr ai-agents architecture architecture-decision-records backstage developer-experience

adr-compliance

Use this agent proactively to audit code changes against accepted Architectural Decision Records. This agent checks if implementation follows documented architectural decisions and flags violations. Examples: Context: User has written code that might conflict with an accepted ADR. user: "I just implemented the new…

zircote-plugins/adr · 266 tokens

decision-checker

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Scope

Method

What this file has done since we first saw it

Other agents, from other repositories

adr-compliance

implementation-strategist

pragmatic-enforcer

domain-expert

maintainability-expert

performance-specialist