## Summary Roll out [the fix for GPT filling every optional MCP tool argument](https://gitlab.com/gitlab-org/gitlab/-/work_items/621520) currently behind the `dap_openai_non_strict_tools` feature flag. When the flag is on for a user, the AI gateway sends Duo Agent Platform tools to OpenAI models with strict mode off, so GPT can leave optional arguments out instead of inventing a value for each one. The flag lives in Rails (gitlab-org/gitlab!261800) and is pushed to the AI gateway with the other Duo Workflow flags; the gateway reads it (gitlab-org/modelops/applied-ml/code-suggestions/ai-assist!7436). Turning it on in Rails does nothing until that gateway change is deployed. - DRI: @terrichu - Team Slack channel: `#g_agent-execution` > [!note] > Process and guidance live in the docs — this issue is just the commands and a place to track the rollout. > "Rolling out" means incrementally enabling the flag on GitLab.com to validate stability — it is not the same as releasing the feature, which happens when the flag is removed. > [Feature flag controls](https://docs.gitlab.com/development/feature_flags/controls/) · [Feature flag lifecycle](https://handbook.gitlab.com/handbook/product-development/how-we-work/product-development-flow/feature-flag-lifecycle/#feature-flag-lifecycle) ## What could go wrong? - Only GPT models are affected; Anthropic and Gemini requests don't change. - With strict mode off, OpenAI no longer guarantees that GPT's arguments match the tool schema. A malformed call fails MCP validation and the model gets a self-correcting error, the same as other models today. - **Why it matters while it's off:** in strict mode GPT sends `milestone_id: 0` on every `save_merge_request` update, and 0 means "remove the milestone" in the REST API, so each of those updates removes the merge request's milestone. That already happens on master. With the flag on, GPT can leave `milestone_id` out. Watch Duo Agent Platform sessions that use GPT models for tool-call validation errors. ## Events - [Exceptions with dap_openai_non_strict_tools:1](https://log.gprd.gitlab.net/app/discover#/?_g=(time:(from:now-1d,to:now))&_a=(index:'7092c4e2-4eb5-46f2-8305-a7da2edad090',query:(language:kuery,query:'json.exception.feature_flag_states.keyword:%22dap_openai_non_strict_tools:1%22'),columns:!(json.exception.class,json.exception.message,json.extra.workflow_id,json.meta.caller_id),sort:!(!(json.time,desc)))) - [Events with dap_openai_non_strict_tools:1](https://log.gprd.gitlab.net/app/discover#/?_g=(time:(from:now-1d,to:now))&_a=(index:'7092c4e2-4eb5-46f2-8305-a7da2edad090',query:(language:kuery,query:'json.feature_flag_states:%22dap_openai_non_strict_tools:1%22'),columns:!(json.meta.caller_id,json.feature_flag_states),sort:!(!(json.time,desc)))) - [Error rate and other graphs by modifying the examples in the Visualization Library](https://log.gprd.gitlab.net/app/visualize#/?_g=h@358d019&s=FF%20Observability) Feature Flag events are only logged by default for feature flags marked for the current or future milestones. To enable while the feature flag is active, see https://docs.gitlab.com/development/feature_flags/#logging ## Rollout Run all production `/chatops` in [`#production`](https://gitlab.slack.com/archives/C101F3796) and cross-post the results to `#g_agent-execution`. Background: [incremental rollout process](https://docs.gitlab.com/development/feature_flags/controls/#process), [feature actors](https://docs.gitlab.com/development/feature_flags/#feature-actors). The flag is checked against the user (`Gitlab::AiGateway.push_feature_flag(:dap_openai_non_strict_tools, current_user)`). **Non-production** ``` /chatops gitlab run feature set dap_openai_non_strict_tools 50 --actors --dev --pre --staging --staging-ref /chatops gitlab run feature set dap_openai_non_strict_tools true --dev --pre --staging --staging-ref ``` **Production** — percentage rollout (wait ≥15 min between steps, watch dashboards): ``` /chatops gitlab run feature set dap_openai_non_strict_tools <percentage> --actors ``` Or target specific actors instead: ``` /chatops gitlab run feature set --user=fdegier dap_openai_non_strict_tools true ``` ## Before global rollout Confirm the relevant gotchas before going to 100% — see [enabling a feature for GitLab.com](https://docs.gitlab.com/development/feature_flags/controls/#enabling-a-feature-for-gitlabcom): - [Docs + version history](https://docs.gitlab.com/development/documentation/feature_flags/) updated - [Breaking changes](https://docs.gitlab.com/development/documentation/release_notes/#deprecations-removals-and-breaking-changes) announced, if any - [Change management issue](https://handbook.gitlab.com/handbook/engineering/infrastructure-platforms/change-management/#feature-flags-and-the-change-management-process) opened, if required - [External API consumers](https://docs.gitlab.com/development/feature_flags/#do-not-use-feature-flags-in-external-api-consumers) handled with a fail-open mechanism, if applicable ## Cleanup Remove the flag once [deemed stable](https://handbook.gitlab.com/handbook/product-development/how-we-work/product-development-flow/feature-flag-lifecycle/#feature-flag-lifecycle) — see [cleaning up](https://docs.gitlab.com/development/feature_flags/controls/#cleaning-up). Remove the flag check in the AI gateway first, then the flag and its YAML definition from Rails, then: ``` /chatops gitlab run release check https://gitlab.com/gitlab-org/gitlab/-/merge_requests/261800 19.6 /chatops gitlab run feature delete dap_openai_non_strict_tools --dev --pre --staging --staging-ref --production ``` ## Rollback ``` /chatops gitlab run feature set dap_openai_non_strict_tools false # production /chatops gitlab run feature set dap_openai_non_strict_tools false --dev --pre --staging --staging-ref # non-production /chatops gitlab run feature delete dap_openai_non_strict_tools --dev --pre --staging --staging-ref --production # remove entirely ```