cm-decider

cm-decider is an agent for Claude Code from Aleksacom/claude-code-orchestration-agents. It costs 60 tokens per session (1,173 once invoked), scanned A, original, MIT.

A read-only planning assistant that decides which open planning questions can be answered automatically and which need a person’s decision. It is used when a planner has several questions.

In plain words
What is it for?
Use it to route planning questions to automatic resolution or human review. It returns decisions about routing, not code or implementations.
Why use it?
It reduces interruptions for routine choices while escalating questions involving scope, public wording, security, pricing, or several parts of a system.

Agent for Claude Code

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 agents/aleksacom/claude-code-orchestration-agents/cm-decider
Clone the repo
git clone --depth 1 https://github.com/Aleksacom/claude-code-orchestration-agents

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 cm-decider

README.md
[![agentmods](https://agentmods.dev/badge/agents/aleksacom/claude-code-orchestration-agents/cm-decider.svg)](https://agentmods.dev/agents/aleksacom/claude-code-orchestration-agents/cm-decider)
Your own site
<a href="https://agentmods.dev/agents/aleksacom/claude-code-orchestration-agents/cm-decider"><img src="https://agentmods.dev/badge/agents/aleksacom/claude-code-orchestration-agents/cm-decider.svg" alt="Measured on agentmods" height="20"></a>
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,173 The whole file, excluding the scripts and references it only reads on demand.
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 5.1 $0.00060 $0.01173
Opus 5 $0.00030 $0.00587
Sonnet 5 $0.00012 $0.00235
Haiku 4.5 $0.00006 $0.00117

Measured 5d ago against content hash c8b4f2649626, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

cm-decider 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 5d 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/agents/cm-decider.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

cm-decider — Question routing authority

You are the cm-decider subagent. You are READ-ONLY. Your job is to evaluate each open question surfaced by cm-planner and route it to either AUTO_RESOLVE (proceed with the recommendation without bothering the user) or ESCALATE (surface to user for explicit decision).

When you're invoked

The main agent calls you when cm-planner has produced 2+ open questions, asking you to triage each one. You read the questions, apply the routing criteria below, and return a verdict per question.

Routing criteria

ESCALATE if ANY of these triggers fire:

# Trigger Examples
1 Planner explicitly hedged on recommendation "could go either way", "depends on...", "unclear"
2 Project identity / user-facing prose tone, copy, public-facing wording
3 Naming that becomes a public API or URL route paths, exported function names, env vars
4 Reframes or supersedes a prior locked decision challenges an existing decision-doc
5 Security, auth, privacy implication secrets, permissions, PII flow
6 Cross-area impact (≥3 systems affected) touches frontend + backend + DB + cache
7 Pricing, payment, financial flow anything touching money
8 First-of-kind in the project new pattern, new library, new domain
9 Paradigm shift in approach changes how the project does X going forward
10 User has explicit say on this category per CLAUDE.md or project rules
11 Reversibility cost is high hard to undo if wrong
12 Convention deviation breaks or extends established patterns
13 Performance trade-off with real impact speed vs accuracy, etc.
14 Citation thin on planner's recommendation reasoning relies on heuristic, not codified rule

AUTO_RESOLVE if:

  • Question is mechanical (file path, conventional naming, formatting)
  • Recommendation is supported by an existing convention or precedent
  • Cost of choosing wrong is low and reversible
  • No ESCALATE trigger fires

Read the full file on GitHub · 101 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. 5d ago First seen · 101 lines · 60 tokens per session scan A c8b4f2649626

Subscribe to this mod's changes

cm-decider is an agent published in the GitHub repository Aleksacom/claude-code-orchestration-agents (2 stars, last pushed 3mo ago), licensed MIT. It adds 60 tokens to every session and 1,173 once invoked, about $0.0003 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.

Related

Other agents, from other repositories

react18-class-surgeon

Class component migration specialist for React 16/17 → 18.3.1. Migrates all three unsafe lifecycle methods with correct semantic replacements (not just UNSAFE prefix). Migrates legacy context to createContext, string refs to React.createRef(), findDOMNode to direct refs, and ReactDOM.render to createRoot. Uses memory…

github/awesome-copilot · 81 tokens

system-architect

Use this agent when making architectural decisions for RTK — adding new filter modules, evaluating command routing changes, designing cross-cutting features (config, tracking, tee), or assessing performance impact of structural changes. Examples: designing a new filter family, evaluating TOML DSL extensions, planning…

rtk-ai/rtk · 0 tokens

Power Platform MCP Integration Expert

Expert in Power Platform custom connector development with MCP integration for Copilot Studio - comprehensive knowledge of schemas, protocols, and integration patterns.

github/awesome-copilot · 31 tokens

MS-SQL Database Administrator

Work with Microsoft SQL Server databases using the MS SQL extension.

github/awesome-copilot · 18 tokens

ap-preflight-probe

L4 diagnostic/recovery probe - on an explicit cache miss, proves RUN/READ/WRITE and reports model/effort bindings; never the mandatory first spawn.

Spielewoy/autoprompt-skill · 39 tokens

ijfw-accessibility-reviewer

Design-phase WCAG 2.1 AA review of UI artefacts: contrast, semantics, focus, ARIA. Trigger per design review pass.

FerroxLabs/ijfw · 37 tokens