github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #13879] Security & Configuration: Granular Feature Control via .env #9080

New Issue
Open
opened 2026-04-12 21:55:23 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 21:55:23 -05:00
Owner
Copy Link

Originally created by @viba1 on GitHub (Jan 24, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/13879

Context:

Since version 0.15.0, the --experimental option allows the execution of system commands via the CLI.
Ollama also offers features like tools, thinking (model reasoning steps), and potentially others in the future.
Currently, controlling these features via the .env file is limited or inexistent.

It is not currently possible to:

  • Prevent the activation of --experimental mode ;
  • Restrict the execution of system commands in experimental mode ;
  • Disable individual features like tools or thinking via configuration.

Problem:

The execution of arbitrary system commands (via --experimental) presents a significant security risk, potentially allowing for destructive operations (e.g., format c:/).
Beyond this, the uncontrolled activation of features like tools and reasoning can introduce instability, unexpected behavior, or resource exhaustion in production environments.
A lack of granular control hinders the secure and reliable deployment of Ollama.

Proposed Solution:

Add configuration options (e.g., in ollama.env) to enable/disable individual features.
This would allow administrators to tailor Ollama's behavior to their specific needs and security requirements.

Examples:

  • OLLAMA_DISABLE_EXPERIMENTAL = true: Disables the execution of experimental mode.
  • OLLAMA_DISABLE_COMMANDS = true: Disables the execution of system commands via the CLI.
  • OLLAMA_DISABLE_TOOLS = true: Disables the use of tools.
  • OLLAMA_DISABLE_THINKING = true: Disables the "thinking" (reasoning steps) feature.

Ideally, these features could be disabled by default and explicitly enabled:

  • OLLAMA_ENABLE_EXPERIMENTAL = true
  • OLLAMA_ENABLE_COMMANDS = true
  • OLLAMA_ENABLE_TOOLS = true
  • OLLAMA_ENABLE_THINKING = true

This provides a flexible and secure approach to managing Ollama's functionality.

Benefits:

  • Enhanced Security: Reduces the potential attack surface and mitigates risks associated with uncontrolled feature usage.
  • Increased Flexibility: Allows administrators to customize Ollama’s behavior for different environments.
  • Improved Stability: Enables disabling potentially unstable or resource-intensive features in production.
  • Simplified Configuration: Provides a centralized and easy-to-manage configuration mechanism.
  • Future-Proofing: A scalable solution that can accommodate new features as Ollama evolves.
Originally created by @viba1 on GitHub (Jan 24, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/13879 **Context:** Since version 0.15.0, the `--experimental` option allows the execution of system commands via the CLI. Ollama also offers features like tools, thinking (model reasoning steps), and potentially others in the future. Currently, controlling these features via the `.env` file is limited or inexistent. It is not currently possible to: * Prevent the activation of `--experimental` mode ; * Restrict the execution of system commands in experimental mode ; * Disable individual features like tools or thinking via configuration. **Problem:** The execution of arbitrary system commands (via `--experimental`) presents a significant security risk, potentially allowing for destructive operations (e.g., `format c:/`). Beyond this, the uncontrolled activation of features like tools and reasoning can introduce instability, unexpected behavior, or resource exhaustion in production environments. A lack of granular control hinders the secure and reliable deployment of Ollama. **Proposed Solution:** Add configuration options (e.g., in `ollama.env`) to enable/disable individual features. This would allow administrators to tailor Ollama's behavior to their specific needs and security requirements. Examples: * `OLLAMA_DISABLE_EXPERIMENTAL = true`: Disables the execution of experimental mode. * `OLLAMA_DISABLE_COMMANDS = true`: Disables the execution of system commands via the CLI. * `OLLAMA_DISABLE_TOOLS = true`: Disables the use of tools. * `OLLAMA_DISABLE_THINKING = true`: Disables the "thinking" (reasoning steps) feature. Ideally, these features could be disabled by default and explicitly enabled: * `OLLAMA_ENABLE_EXPERIMENTAL = true` * `OLLAMA_ENABLE_COMMANDS = true` * `OLLAMA_ENABLE_TOOLS = true` * `OLLAMA_ENABLE_THINKING = true` This provides a flexible and secure approach to managing Ollama's functionality. **Benefits:** * **Enhanced Security:** Reduces the potential attack surface and mitigates risks associated with uncontrolled feature usage. * **Increased Flexibility:** Allows administrators to customize Ollama’s behavior for different environments. * **Improved Stability:** Enables disabling potentially unstable or resource-intensive features in production. * **Simplified Configuration:** Provides a centralized and easy-to-manage configuration mechanism. * **Future-Proofing:** A scalable solution that can accommodate new features as Ollama evolves.
GiteaMirror added the feature request label 2026-04-12 21:55:23 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label feature request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#9080
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?