don

A command that temporarily applies a named masque, meaning a predefined perspective and set of working instructions, to the coding agent.

In plain words
What is it for?
Use it with a masque name and optional task intent to adopt that masque and record it as the active session state.
Why use it?
It lets you switch the agent’s working approach for a task without manually copying the instructions into the conversation.

Command

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 commands/chrisdbaldwin/masques/don
Clone the repo
git clone --depth 1 https://github.com/ChrisDBaldwin/Masques
Per session 16 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,723 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 $0.00016 $0.01723
Opus 5 $0.00008 $0.00861
Sonnet 5 $0.00003 $0.00345
Haiku 4.5 $0.00002 $0.00172

Measured 2d ago against content hash cdbb85273b8c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

don 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 2d 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/don.md · 191 lines

How it starts

The opening of the file, as written. The whole thing — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Don Masque Command

You are donning the masque: $ARGUMENTS

Instructions

Step 1: Parse Arguments

Extract the masque name from arguments:

  • First word is the masque name (case-insensitive)
  • Remaining words (if any) are the user's intent

If no arguments provided: Report an error and exit:

✗ Missing masque name

Usage: /don <masque-name> [intent]

Run /list to see available masques.

Do not proceed to subsequent steps.

Step 2: Compose the identity via the masques-cli CLI (single source of truth)

The masque is resolved and composed by the masques-cli CLI — the one authoritative implementation shared by this plugin and the Masques MCP server (PRD v1.2 M7). Do not re-derive the identity block in prose; that is how the two surfaces used to drift. Run the CLI and inject its output verbatim.

Locate the CLI (the $MASQUE invocation used in this and later steps):

  • If masques-cli is on PATH (installed via uv tool install "${CLAUDE_PLUGIN_ROOT}/services/mcp"), use masques-cli.
  • Otherwise fall back to: uv run --project "${CLAUDE_PLUGIN_ROOT}/services/mcp" masques-cli

Compose the identity block — this both resolves the masque (private ${MASQUES_HOME:-~/.masques} over bundled personas/) and builds the block:

$MASQUE compose <name> [intent...]

The CLI exits non-zero and prints a diagnostic to stderr if the masque is not found, the YAML is malformed, or a required field (name, version, lens) is missing. If it fails, report that diagnostic and stop — do not proceed to subsequent steps. For a not-found error, also suggest /list.

To capture the masque's name, version, source, and spinnerVerbs for the steps below (session file, spinner verbs), additionally run:

$MASQUE compose <name> --json

Step 4: Inject the Masque Context

Inject the stdout of $MASQUE compose <name> [intent...] verbatim — it is already the <masque-active> block (lens + context + attributes + optional intent). Do not reformat or regenerate it; emitting it verbatim is what guarantees the plugin and the MCP server compose identically.

Read the full file on GitHub · 191 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. 2d ago First seen · 191 lines · 16 tokens per session scan A cdbb85273b8c

Subscribe to this mod's changes

don is a command published in the GitHub repository ChrisDBaldwin/Masques (14 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 16 tokens to every session and 1,723 once invoked, about $0.0001 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.