The reader-preamble prompt

The whole text of config/agent-prompts/reader-preamble.md, as an agent is handed it.

Bundled agent prompts — one row of 23, exported from config/agent-prompts · every prompt · the resolution order

What this prompt is

id	reader-preamble	the key every tier, every pattern step and every override spells it by
kind	not a stage instruction	whether any bundled pattern hands it to an agent step
role	—	the harness role the agent reading it runs under
stage	—	the stage of a run it is handed at
bundled at	config/agent-prompts/reader-preamble.md	where the built-in copy sits in Gantry's own tree

No bundled pattern hands this file to an agent step.

Where an override of it goes

tier	the file to write	what it is
run consulted 1st	`.gantry/<plan>/prompts/reader-preamble.md`	the run's own frozen copy, written when the run started and read by every agent it launches; editing it changes what a run's remaining agents are told
project consulted 2nd	`.gantry/<plan>/reader-preamble.md`	a per-run override an operator places beside the run's other orchestration files
global consulted 3rd	`~/.config/gantry/agent-prompts/reader-preamble.md`	the materialized global mirror; it overrides the built-in once its bytes differ from the built-in's
built-in consulted 4th	`config/agent-prompts/reader-preamble.md`	the default compiled into the binary

The text

config/agent-prompts/reader-preamble.md

Reader-facing Gantry cards are for an operator who needs to understand run state quickly.
Use Gantry vocabulary carefully: a task is one piece of agent work, the gate is the test script Gantry runs,
a stop is the reason automation paused, and provenance is only evidence present in the supplied input.
Name files, commands, flags, and identifiers exactly as the material spells them — no metaphor or synonym where the literal term exists — and cut every sentence that carries no fact.
Write plainly, name concrete next actions when asked, and never imply facts that are not in the supplied material.