agentry/voice-writer
Writes a Writing Voice: reads samples of a person's writing and writes VOICE.md, how an LLM should write the way they do.
Built into Agentry. The application runs this package itself; it is published here so you can read how it works. Its README says what Agentry gives it beyond its manifest.
Declared capabilities
Capabilities are the author's declaration and the basis for policy and scanning. The registry cannot enforce them at run time. Your runtime should.
MCP servers & tools
agentry run serves exactly the tools this contract declares, through a per-run MCP gateway. An undeclared capability is a tool that does not exist at run time.
Tools the runner grants
| Manifest declaration | Tool(s) granted |
|---|---|
| (always) | write_report — reports → the run's output folder |
| (always) | list_inputs, read_input, search_inputs — operator-provided input files, read-only |
MCP servers involved
- QuickAgent Tools — the runner's companion server, served through a per-run Docker MCP Gateway; its tool surface is exactly the table above.
- Operator-cataloged MCP servers may be attached for the declared dataAccess category (files); undeclared categories are never wired in.
In an embedded install
Installed into an existing coding agent (agentry install), the same contract maps to the host's built-in tools — enforcement there is the host's. Allowed tools: none declared (host defaults apply).
Install
agentry install agentry/voice-writer@1.1.0 --target=agents-mdgit clone https://secureagentry.net/agentry/voice-writer.gitgit clone ssh://git@git.secureagentry.net/agentry/voice-writer.gitagents-md is not this agent's default; the flag is required.
- agent.yaml499 B
- INSTRUCTION.md5.1 kB
- LICENSE1.1 kB
- README.md2.2 kB
README.md
The Voice Writer
The Voice Writer is the QuickAgent that writes a person's Writing Voice.
You give it samples of your own writing, and it writes VOICE.md: the
instructions an LLM follows to write the way you do. With a voice attached,
a QuickAgent run writes its answers, reports and messages in it. Run it
again with more samples and a brief to improve the voice, or with a note
(and, if you like, more samples) to edit it.
This is the package that runs
The Voice Writer is built into the Agentry CLI and the desktop app, and it
runs from there: on your computer when you run agentry voice create,
improve or edit (or make a voice in the desktop), and on an Agentry
sandbox host when you make one in Agentry. This release has the same bytes.
The registry built its archive from the package the CLI carries, and the
archive's digest is the one the CLI reports. Run agentry builtins to print
it and compare it with the digest on this page.
Installing this release does not give you the Voice Writer. The application runs it as a built-in, with the steps and tools listed below. If you install this release, it runs as an ordinary QuickAgent with only what its manifest declares.
What the application does around it
The manifest declares no network access, no file access and no commands, and
one data category, files. When Agentry writes a voice:
- Before the run, the runner decodes each sample to text: a PDF, an
office file or an image becomes a
.txtfile, and a text file stays as it is. The Voice Writer runs no decoder itself. It reads the text, and the current voice when there is one, through the inputs channel (list_inputs,read_input). - During the run, it writes
VOICE.mdwithwrite_report. - After the run, the application checks the file's form (UTF-8 text, at most 32 KiB), not what it says, and keeps it as the voice. A voice made on your computer stays there and is not reviewed. A voice that reaches your Agentry account is reviewed before it is used there, to check that it holds only writing instructions.
INSTRUCTION.md is the instruction the model follows. It covers what a
voice says, what it never says, and how the samples are read.