psychology-agent: Command for Claude Code

.claude/commands/adjudicate.md

adjudicate is a command for Claude Code from safety-quotient-lab/psychology-agent. It costs 0 tokens per session (1,959 once invoked), scanned A, original, Apache-2.0.

A structured decision command for comparing two or more options by tracing their effects, comparing them across relevant factors, and choosing a resolution. Its shorter mode handles small decisions; its full mode examines larger ones in two passes.

In plain words
What is it for?
Use it to decide between implementation or design options, analyze each option's knock-on effects, compare trade-offs, and settle on a consensus or the simplest defensible choice.
Why use it?
It helps make ambiguous choices explicit and exposes consequences that may appear several steps later.

Command for Claude Code

Written for Claude Code: $ARGUMENTS substitution.

This is safety-quotient-lab/psychology-agent's own configuration. It tells Claude Code how to work on psychology-agent 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 psychology-agent configures →

Reuse

Borrowing it

Nothing to install: this file belongs to safety-quotient-lab/psychology-agent. 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/safety-quotient-lab/psychology-agent/main/.claude/commands/adjudicate.md
Clone the repo
git clone --depth 1 https://github.com/safety-quotient-lab/psychology-agent

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 adjudicate

README.md
[![agentmods](https://agentmods.dev/badge/commands/safety-quotient-lab/psychology-agent/adjudicate/github.svg)](https://agentmods.dev/commands/safety-quotient-lab/psychology-agent/adjudicate)
Your own site
<a href="https://agentmods.dev/commands/safety-quotient-lab/psychology-agent/adjudicate"><img src="https://agentmods.dev/badge/commands/safety-quotient-lab/psychology-agent/adjudicate/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 adjudicate

Your own site · 80×15
<a href="https://agentmods.dev/commands/safety-quotient-lab/psychology-agent/adjudicate"><img src="https://agentmods.dev/badge/commands/safety-quotient-lab/psychology-agent/adjudicate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,959 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 $0.00000 $0.01959
Opus 5 $0.00000 $0.00979
Sonnet 5 $0.00000 $0.00392
Haiku 4.5 $0.00000 $0.00196

Measured 11d ago against content hash 2cded874b405, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

adjudicate 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 11d 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/commands/adjudicate.md · 237 lines

How it starts

The opening of the file, as written. The whole thing — 237 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Adjudicate — Structured Decision Resolution

Resolve ambiguous decisions with 2+ options through iterative knock-on analysis, multi-axis comparison, and consensus-or-parsimony resolution.

Adjudication is the full decision-making unit: it composes /knock (single- option effect tracing) with iterative learning (2-pass), structured comparison (differentiating axes), and binding resolution (consensus or parsimony).

/knock       = trace effects of ONE option/change through 10 orders
/adjudicate  = 2-pass /knock on EACH option → compare → resolve

Arguments

Parse $ARGUMENTS to determine scope:

Argument Behavior
(empty) Identify the decision from context, adjudicate
[topic] Adjudicate the named decision
inline Abbreviated: 4-order, no 2-pass, for S decisions
full Full protocol: 10-order, 2-pass, for M/L decisions

If scale is not specified, infer from context:

  • XS/S decisions → inline (4-order knock-on + parsimony + structural scan)
  • M/L decisions → full (10-order, 2-pass, comparison, consensus/parsimony)

Protocol

Phase 1: Frame the Decision

  1. State the decision clearly in one sentence
  2. Identify 2-3 mutually exclusive options (never more than 4)
  3. For each option, state it as a concrete action, not a vague direction
  4. Classify: is this XS, S, M, or L effort/impact?

Phase 2: Knock-on Analysis (2-pass for M/L)

For each option, trace effects through 10 orders:

 Order   Confidence           What it captures
─────────────────────────────────────────────────────────────────
 1       certain              Direct, immediate effect
 2       certain–likely       What systems/processes activate
 3       likely               What consumes Order 2's outputs
 4       likely–possible      Aggregate/systemic effects
 5       possible             What humans observe / trust changes
 6       speculative          How it compounds over time
 7       structural           Ecosystem/precedent effects —
                              how does this shape what others do?
                              What precedent for open-source,
                              community, or standards?
 8       horizon              Normative/structural effects —
                              what norms, expectations, or
                              constraints does this establish
                              or erode long-term?
 9       emergent             Properties arising from interaction
                              of multiple knock-on chains — not
                              predictable from individual orders
                              in isolation (INCOSE)
 10      theory-revising      Effects that falsify or require
                              modification of the theory that
                              justified the original decision
                              (Popper)
─────────────────────────────────────────────────────────────────

Read the full file on GitHub · 237 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. 11d ago First seen · 237 lines · 0 tokens per session scan A 2cded874b405

Subscribe to this mod's changes

adjudicate is a command published in the GitHub repository safety-quotient-lab/psychology-agent (20 stars, last pushed 4mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,959 tokens. 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.