Go good first issues

AI Summary: This issue proposes adding an inline viewer for text files within the existing pod filesystem browser. Currently, all files are forced to download, preventing quick inspection of small configuration or log files. The goal is to allow users to view supported text files directly in the UI, with clear fallbacks for unsupported or oversized files, while maintaining existing download functionality and RBAC.

AI Summary: This issue proposes adding file import and drag-and-drop functionality to the Radar application for applying multi-resource YAML files. Currently, users can only paste YAML into an editor, making it cumbersome to apply larger manifest bundles. The goal is to integrate this new functionality into the existing resource creation/apply dialog, reusing the current apply pipeline for a seamless user experience.

AI Summary: The OpenAPI specification incorrectly declares the `models` array in the `/api/inference/models/{backend}` endpoint as containing objects, while the Go client expects strings. This mismatch causes errors when the client attempts to unmarshal the response. The issue highlights a gap in testing, as existing tests only verify route coverage, not the accuracy of declared data shapes.

AI Summary: The `ci-maintainer` agent is permanently paused in all governor modes because its cadence key defaults to an agent that is never enabled, causing the cadence to fall back to 0 (paused). This issue requires a decision on the appropriate cadence for `ci-maintainer` in different modes and whether the fallback mechanism for missing cadences is appropriate.

AI Summary: This issue identifies a significant number of test failures in the `pkg/dashboard` package that are only exposed when tests are run in a shuffled order, not by repeated runs. The failures stem from shared mutable state within the package that is not properly reset between tests, leading to order-dependent behavior and potential race conditions. The suggested approach is to fix these issues incrementally, addressing one shared global state per pull request.

AI Summary: A bug in the analysis tool causes it to incorrectly inflate the nesting depth of flat `if/elif` chains. Each `elif` arm is counted as a separate nesting level, leading to artificially high nesting depths and potentially misclassified high-risk functions. The expected behavior is to count only actual nested structures, not sequential `elif` clauses.

AI Summary: The `pkg/dashboard` package has become a monolithic 'god package' by directly importing and depending on 32 other internal packages, including major subsystems. This high coupling leads to widespread impact from changes, slow compilation and testing, and prevents effective mocking or reuse of the dashboard. The recommendation is to introduce dependency inversion by defining interfaces at the dashboard's boundary and wiring concrete implementations at the composition root.

AI Summary: The `saas.go` file in `pkg/hub` has grown excessively large, containing over 22,000 lines of code and numerous unrelated functions. This monolithic file leads to frequent merge conflicts, hinders code reviews due to tooling limitations, and obscures the true dependency structure. The recommendation is to split this file into smaller, more focused files based on their respective concerns.

AI Summary: The escalation reviewer lane in v5 of the project is not functional because its work-list builder function, `buildReviewerWorkList`, has no callers outside its own file. This means the Go half of the escalation-adjudication lane is present but unreachable, leading to a feature that is merged but not wired.

AI Summary: This issue identifies a bug where the cgroup path is hardcoded in nerdctl, causing problems specifically on WSL2.9 and later versions. The hardcoded path needs to be dynamically determined to ensure compatibility with the changing cgroup structures in newer WSL2 environments. The project is seeking assistance from WSL2 users to resolve this.

AI Summary: This issue highlights a critical gap in RFC #4002, which proposes a significant architectural change to the agent turn model. The RFC lacks a phased delivery roadmap, compatibility constraints, and a clear version target. Without a plan, there's a risk of partial implementation, breaking changes without migration guidance, and an undecided performance/durability tradeoff.

AI Summary: Upgrading to v5 of the software introduces a new UID-isolation feature that causes startup failures on existing data volumes. This is a critical blocker for zero-downtime upgrades as the new feature iterates over the entire data PVC on every start. There is currently no migration guide or documentation for this change, leaving operators unable to safely upgrade.

AI Summary: This issue proposes adding a server-side field selector to the Service informer to filter out headless Services. This optimization aims to reduce cache memory, apiserver bandwidth, and processing overhead by preventing the informer from watching, caching, or listing Services that lack a ClusterIP, as these are already ignored by downstream consumers. The change leverages a pattern adopted in Kubernetes 1.31 and requires a compatible apiserver.

AI Summary: The TTFT load gate incorrectly compares observed endpoint TTFT values against substituted defaults when an endpoint lacks the necessary load attribute. This leads to unexpected behavior where the gate might either lose affinity or hold it under arbitrary overload, depending on the `ttftSource` and which endpoints are missing the attribute. The issue is particularly relevant when using the `latencyPredictor` source, which is reachable in production.

AI Summary: The GitHub Actions workflow file `backend-smoke.yml` is failing on every push due to an invalid use of the `matrix` context within a job-level `if` condition. This prevents all subsequent runs, including scheduled ones, from executing. The recommended fix is to remove the `matrix.lane` reference from the job-level `if` and apply the filtering within a `scope` step where the `matrix` context is accessible.

AI Summary: This issue details three bugs found in the `kick-governor.sh` script. First, pause checks are ignored when `HIVE_REPOS` is already set, leading to ignored pauses. Second, the script crashes under budget pressure on fleets missing specific agents like `outreach` or `supervisor`. Finally, a downgrade target for a non-existent `haiku` model version is being generated.

AI Summary: This bug report describes a scenario where COSI can become stuck during the deletion of a Bucket and its associated BucketClaim. This occurs if the Bucket is deleted before the BucketClaim deletion process has fully reconciled, leading to a deadlock where neither resource can complete its deletion due to missing or conflicting annotations.

AI Summary: This issue identifies a gap in CI testing coverage for `src/deploy/` test suites, similar to a previously addressed issue for `bin/` test suites. Currently, 30 `src/deploy/test_*.sh` suites exist, but five are not wired into any CI workflow, with one suite, `test_entrypoint_symlinks.sh`, running nowhere. This is attributed to individual wiring of tests, leading to omissions going unnoticed. The proposal is to extend the existing CI guard or create a new one to ensure all `src/deploy/test_*.sh` suites are referenced in CI workflows.

AI Summary: The script `bin/enumerate-actionable.sh` fails to re-check SHAs for external issues and resolve closed-issue markers because `xargs -I {}` incorrectly substitutes the placeholder `{}` within Python dictionary literals in the executed command. This leads to `SyntaxError`s in the child script, which are then silently masked as 'skip'. The fix involves changing how `xargs` handles arguments to prevent this unintended substitution.

AI Summary: The bot's comment finding function only checks the first 50 comments, preventing it from editing its own previous digest on issues with more than 50 comments. This leads to the bot creating new comments instead of editing existing ones, even after a related bug fix. The suggested solution is to implement pagination in the comment finding function.

AI Summary: This issue aims to remove outdated and unprofiled end-to-end test cases related to 'rl-driven' selections. The work involves deleting specific test files, removing references to these tests, and adding a new test to ensure they are no longer registered. The goal is to clean up the test suite and prevent contributors from being directed to unmaintained tests.

AI Summary: The system silently accepts extremely low token budget limits (e.g., 5, 50, 1000 tokens) during configuration loading and saving, leading to spokes that are permanently budget-exhausted and indistinguishable from other inactive spokes. The proposed solution involves implementing a sanity floor for budget limits with prominent warnings and a distinct health state for budget-exhausted spokes.

AI Summary: The governor component is repeatedly posting identical advisory digest comments to the same GitHub issue, even when there are no new findings. This creates excessive noise for subscribers, consumes API quota, and buries important updates. The proposed solution involves implementing content-based hashing to detect and suppress duplicate posts, as well as a hard cap on consecutive zero-finding digests per target issue.

AI Summary: The main hive process fails to reload `beads.json` due to incorrect file permissions (0600) set by the agent user, while the main process runs with different user privileges. The proposed fix involves ensuring files are written with group-readable permissions and implementing a self-healing mechanism in the load/reload path to correct file modes when permission errors occur.

AI Summary: The visualization of the CVSS distribution trend is not displaying correctly, as indicated by the provided screenshot. This issue appears to be affecting the devguard-web component and has been labeled as a high priority.

AI Summary: The user wants to reduce the font size of headlines on the dashboard to match the styling of the "CVSS Trend" section. This is a visual adjustment to improve consistency and potentially readability.

AI Summary: This issue aims to consolidate image metadata (Dockerfile, context, platform) into a single catalog to prevent drift between the CI validation and publishing workflows. The work involves refactoring duplicated logic into a shared script or data file and updating policy validation to reference this new catalog. The goal is to ensure consistency and simplify maintenance when adding or renaming images.

AI Summary: This issue aims to consolidate the setup process for Kind and disk space management in both the Operator and E2E CI workflows. By creating a shared bootstrap action, the goal is to ensure consistency, reduce duplication, and prevent divergence between these critical CI jobs.

AI Summary: This issue proposes to consolidate Go toolchain setup in CI by creating a shared composite action for Go 1.25. This will eliminate redundant installations and executions of core toolchain steps across different CI jobs, specifically in PR workflows and `operator-ci.yml`. Additionally, it aims to remove unnecessary `golangci-lint` and `go-mod-tidy` checks that are already covered elsewhere.

AI Summary: This issue highlights two critical discoverability gaps for the project: the absence of relevant GitHub repository topics and a ROADMAP.md file. These omissions prevent new visitors from easily finding and understanding the project's direction and potential, acting as an adoption blocker. The proposed next steps involve adding appropriate topics and creating a comprehensive roadmap document linked from the README.