anomalyco/opencode
1
0
Fork 0
mirror of https://github.com/anomalyco/opencode.git synced 2026-07-21 17:56:33 -04:00
Code Issues 5.9k Packages Projects Releases 19 Wiki Activity

[FEATURE]: Requires secondary model validation for tools that produce side effects. #8522

New Issue
Open
opened 2026-02-16 18:10:11 -05:00 by yindo · 2 comments
yindo commented 2026-02-16 18:10:11 -05:00
Owner
Copy Link
Copy Source

Originally created by @RonxBulld on GitHub (Feb 4, 2026).

Originally assigned to: @thdxr on GitHub.

Feature hasn't been suggested before.

  • I have verified this feature I'm about to request hasn't been suggested before.

Describe the enhancement you want to request

As the title suggests, tools that produce side effects require secondary model validation. When a model calls a tool that might alter the content of any object, a prompt word should be injected to force the model to reconsider and require careful verification.

I currently have a preliminary implementation; if you agree, I'd be happy to create a PullRequest.

Originally created by @RonxBulld on GitHub (Feb 4, 2026). Originally assigned to: @thdxr on GitHub. ### Feature hasn't been suggested before. - [x] I have verified this feature I'm about to request hasn't been suggested before. ### Describe the enhancement you want to request As the title suggests, tools that produce side effects require secondary model validation. When a model calls a tool that might alter the content of any object, a prompt word should be injected to force the model to reconsider and require careful verification. I currently have a preliminary implementation; if you agree, I'd be happy to create a PullRequest.
yindo added the discussion label 2026-02-16 18:10:11 -05:00
yindo commented 2026-02-16 18:10:11 -05:00
Author
Owner
Copy Link
Copy Source

@github-actions[bot] commented on GitHub (Feb 4, 2026):

This issue might be a duplicate of existing issues. Please check:

  • #10238: Safer default for build agent: prompt on destructive bash commands
  • #10820: Add metadata awareness and data type validation/blacklists for LLM tool execution
  • #11382: Feature Proposal: Enhance Core bash Tool with Sessions & Safety Guardrails

These issues address related concerns about validating tool execution, preventing destructive operations, and adding safety guardrails for tool calls. While your proposal specifically focuses on injecting a prompt to force model reconsideration for side-effect-producing tools, the above issues discuss similar safety mechanisms and validation approaches.

Feel free to ignore if your specific implementation approach is different from these.

@github-actions[bot] commented on GitHub (Feb 4, 2026): This issue might be a duplicate of existing issues. Please check: - #10238: Safer default for build agent: prompt on destructive bash commands - #10820: Add metadata awareness and data type validation/blacklists for LLM tool execution - #11382: Feature Proposal: Enhance Core bash Tool with Sessions & Safety Guardrails These issues address related concerns about validating tool execution, preventing destructive operations, and adding safety guardrails for tool calls. While your proposal specifically focuses on injecting a prompt to force model reconsideration for side-effect-producing tools, the above issues discuss similar safety mechanisms and validation approaches. Feel free to ignore if your specific implementation approach is different from these.
yindo commented 2026-02-16 18:10:11 -05:00
Author
Owner
Copy Link
Copy Source

@RonxBulld commented on GitHub (Feb 4, 2026):

This issue might be a duplicate of existing issues. Please check:

* [[FEATURE]: Safer default for build agent: prompt on destructive bash commands #10238](https://github.com/anomalyco/opencode/issues/10238): Safer default for build agent: prompt on destructive bash commands

* [Add metadata awareness and data type validation/blacklists for LLM tool execution #10820](https://github.com/anomalyco/opencode/issues/10820): Add metadata awareness and data type validation/blacklists for LLM tool execution

* [Feature Proposal: Enhance Core bash Tool with Sessions & Safety Guardrails #11382](https://github.com/anomalyco/opencode/issues/11382): Feature Proposal: Enhance Core bash Tool with Sessions & Safety Guardrails

These issues address related concerns about validating tool execution, preventing destructive operations, and adding safety guardrails for tool calls. While your proposal specifically focuses on injecting a prompt to force model reconsideration for side-effect-producing tools, the above issues discuss similar safety mechanisms and validation approaches.

Feel free to ignore if your specific implementation approach is different from these.

The proposals in the aforementioned threads are fundamentally different from this proposal.

@RonxBulld commented on GitHub (Feb 4, 2026): > This issue might be a duplicate of existing issues. Please check: > > * [[FEATURE]: Safer default for build agent: prompt on destructive bash commands #10238](https://github.com/anomalyco/opencode/issues/10238): Safer default for build agent: prompt on destructive bash commands > > * [Add metadata awareness and data type validation/blacklists for LLM tool execution #10820](https://github.com/anomalyco/opencode/issues/10820): Add metadata awareness and data type validation/blacklists for LLM tool execution > > * [Feature Proposal: Enhance Core bash Tool with Sessions & Safety Guardrails #11382](https://github.com/anomalyco/opencode/issues/11382): Feature Proposal: Enhance Core bash Tool with Sessions & Safety Guardrails > > > These issues address related concerns about validating tool execution, preventing destructive operations, and adding safety guardrails for tool calls. While your proposal specifically focuses on injecting a prompt to force model reconsideration for side-effect-producing tools, the above issues discuss similar safety mechanisms and validation approaches. > > Feel free to ignore if your specific implementation approach is different from these. The proposals in the aforementioned threads are fundamentally different from this proposal.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
acp
bad
bug
discussion
docs
documentation
enhancement
extension
good first issue
growth
help-wanted
mcp
model-problem
needs:compliance
nix
opentui
perf
pull-request
Mirrored from GitHub Pull Request
 tui
web
windows
zen
No Label discussion
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: anomalyco/opencode#8522
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?