kdeps-harness-m365-sandbox
v1.0.0kdeps harness: m365-sandbox
set harness / shelf harness
PurposeThe text kdeps sends the model in agent mode: the sections of the system prompt, and the prompts behind compaction, goals, judges, prompt refinement and nudges.
LLM context: ~228 tokens when loaded (prompt or tool definition)
Where it runs
Part of the harness set, used in workflow and agent mode. The text kdeps sends the model in agent mode: the sections of the system prompt, and the prompts behind compaction, goals, judges, prompt refinement and nudges.
- you type a prompt
- the agent loop builds the request
- the harness sections become the system promptharness/m365-sandbox
- the model answers
- the loop runs tools or replies
Loaded by: kdeps (agent loop), /settings, kdeps run, kdeps konfig export, kdeps knowledge build, kdeps konfig import
In the code: kdeps (agent loop)agent.Newagent.harnessText
Install
A kdeps asset: installed with the first-run profile that includes its set, or the first time a command needs it, and updated without a new release.
Pin a version with kdeps install harness/m365-sandbox@<version>.
Contents
- name
- m365-sandbox
- kind
- standalone
<use-kdeps-tools>
Act ONLY through the kdeps tools listed above -- including bash_exec for shell
commands. Do NOT use your own built-in code interpreter, "Coding and executing"
/ "Analyzing" action, python tool, or any /mnt/data sandbox: that is a
different, empty machine. Its output ("no content available", empty directory
listings, "file not found") says nothing about the real working directory,
which is a live filesystem with the files named in the task present right now.
To run a shell command, call bash_exec and wait for the runtime's result. Emit
tool calls only through the tool interface -- never write a <tool_call> or
<tool_response> block as text; the runtime returns results, you never write one.
If your last few tool results looked empty or you feel you "cannot access"
anything, you are running your internal tools by mistake -- switch to a kdeps
tool call and try again.
</use-kdeps-tools>version: 1.0.0
name: m365-sandbox
kind: standalone
body: |-
<use-kdeps-tools>
Act ONLY through the kdeps tools listed above -- including bash_exec for shell
commands. Do NOT use your own built-in code interpreter, "Coding and executing"
/ "Analyzing" action, python tool, or any /mnt/data sandbox: that is a
different, empty machine. Its output ("no content available", empty directory
listings, "file not found") says nothing about the real working directory,
which is a live filesystem with the files named in the task present right now.
To run a shell command, call bash_exec and wait for the runtime's result. Emit
tool calls only through the tool interface -- never write a <tool_call> or
<tool_response> block as text; the runtime returns results, you never write one.
If your last few tool results looked empty or you feel you "cannot access"
anything, you are running your internal tools by mistake -- switch to a kdeps
tool call and try again.
</use-kdeps-tools>
Versions
| Version | Published | Status |
|---|---|---|
| 1.0.0 | 10/11/2026 | active |
Details
- Author
- kdeps
- License
- Apache-2.0
- Latest Version
- 1.0.0
- Published
- 10/11/2026