JevMade field notes / Technical guide

Use Jev answers in command-line automation

SemDecide lets small automation programs ask Jev about text. It returns predictable answers and separate status numbers so uncertainty is not mistaken for yes, no, or success.

Original by sharzikiAgent workflows

Listen to this guide

JevMade’s plain-English explanation

0:00 /

AI narration

Credits and license

“SemDecide” by sharziki. Read the original source.

This expanded guide is an AI-narrated adaptation of the source’s essential explanation, examples and caveats, not a word-for-word reading. The synthetic voice does not imitate the author or imply their endorsement.

License: MIT

MIT License

Copyright (c) 2026 Sharvil Saxena

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Our summary

Command-line automation joins small programs that pass information between them. SemDecide lets those programs ask Jev fixed questions about text or records. It can return yes or no, choose a named route, rate described levels, or select records to pass onward.

Different status numbers identify no, uncertainty, invalid input, and service failure. Local limits control text size, record counts, repeated attempts, and waiting time. A checking recipe asks separate hazard questions, then ordinary software combines the answers into allow, ask for help, or block decisions.

The source warns that this does not grant permission or isolate dangerous programs. Submitted text goes to TypeSafe, and the judgment can be wrong. Money, private data, secret keys, live services, and irreversible actions still need exact checks and cautious responses to failures.

Key takeaways

Give uncertain answers their own outcome.
Limit input, repeated attempts, and waiting time.
A model judgment does not grant permission to perform an action.

More guides →See the experiments →