Go good first issues

This issue involves adding a new configuration block for a "bead synthesizer" within the knowledge publishing system. The focus is on implementing the configuration structure and API endpoints for loading and saving these settings, without implementing the actual publishing logic. This includes defining default values, validation rules, and ensuring owner-only access to the API.

The 'Work balance' panel in Bluefin currently only displays issues from the hive audit log, leading to an incomplete view when compared to the project's thousands of GitHub issues. This issue proposes creating a GitHub-sourced weekly issue series to backfill this data, allowing the 'Work balance' panel to show a more comprehensive picture by merging audit and GitHub issue counts.

The frontmatter parser incorrectly handles single-quoted YAML values containing apostrophes. When an apostrophe appears within a single-quoted string, it's escaped by doubling it in YAML. The current parser strips the outer quotes but leaves the doubled apostrophe, resulting in an incorrect literal value. This needs to be fixed so that doubled apostrophes are correctly interpreted as a single apostrophe.

This issue proposes to retire the `stage/regressed` label and redefine `stage/stalled` to indicate a lack of activity on a PR. The current labels are causing confusion by flagging normal review processes as regressions or stalls, even when CI is green or reviews are pending. The proposed changes aim to make these labels more accurately reflect the state of a PR and reduce unnecessary alarm.

The Experiment Lab's Suggest mode fails to produce proposals because the 'hive' binary lacks an in-process proposal generator. Proposals are currently only generated by an external experiment loop, which is not present in a stock installation. This issue requires either shipping an in-process proposal designer or ensuring the external experiment loop is deployed.

This issue requests the implementation of a new Git provider for Bitbucket. It specifically mentions building upon the new structural changes introduced in a recent pull request, indicating a need to integrate with an existing framework.

This issue proposes adding a new configuration setting to the Burrito tool. This setting would prevent Burrito from executing 'apply' operations if the 'plan' output indicates that any resources are scheduled for destruction. The goal is to enhance safety and facilitate adoption on critical Terraform code by preventing accidental resource deletion.

A lint check automatically renames files by slugifying them, but this process incorrectly removes special characters like ß, æ, œ, þ, and ij. The fix involves updating the slugification logic to correctly handle these characters while ensuring the output still conforms to the expected slug pattern.

This issue addresses a bug where the `--help` flag for several Cartographer CLI commands (`kb`, `client`, and `workspace`) incorrectly prints usage information to stderr and exits with an error code. The goal is to modify these commands to print usage to stdout and exit successfully (code 0) when `--help` is provided, mirroring the behavior of other commands like `paths`, `service`, and `audit`. This fix should also ensure that `client` and `workspace` commands can display help even without a configuration file.

This issue requests adding a `--json` flag to the `cartographer kb list` command to provide machine-readable output for scripts and agents. The implementation should follow the pattern of the existing `kb repair --json` command and ensure the output format matches the specified JSON structure, including handling cases where the server doesn't respond.

This issue proposes an enhancement to the Cartographer CLI's `resolve` command. When a path placeholder cannot be resolved, the error message should be improved to include the two commands that can fix the issue: `cartographer paths set` and `cartographer paths suggest`. Additionally, the command's argument parsing should be refactored to use an existing helper function.

This issue requires updating documentation and configuration for a knowledge curator feature, specifically documenting its fields and a PR-only publishing policy. Additionally, it involves a minor code change to include a specific vault in spoke backups, even when publishing is disabled, to ensure its data is preserved.

The `hive/claimed-by-<who>` label on GitHub issues is inconsistently applied, sometimes using the agent's backend name (like 'copilot') instead of the actual human's GitHub login. This issue proposes a unified approach to derive the label from the claim marker, ensuring consistency for both human and agent claims and fixing existing inconsistencies.

This issue proposes adding functionality to the bead synthesizer to publish its generated knowledge facts directly to a Git repository. Currently, these facts are only written to a non-durable PVC, making them susceptible to loss and inaccessible for review in Git. The proposed solution involves an owner-only setting in the dashboard to configure repository publishing, including options for the target repository, path, branch, commit mode, cadence, and update semantics.

This issue aims to add the median time-to-merge metric to the evidence pipeline, specifically for the hub drill-down. It involves modifying spoke and hub components to collect and aggregate this data, updating the UI to display it, and documenting the changes. The implementation requires careful handling of missing data and relies on a prerequisite PR that adds the necessary `merge.opened_at` field.

This issue involves building a "bead-synth publisher" that renders synthesized facts into markdown files and publishes them to a repository via pull requests. It needs to handle updates, avoid duplicates, and skip sensitive or filtered facts. The publisher will integrate with existing knowledge modules and be tested with a fake GitHub client.

This RFC proposes an "evidence pipeline" to measure the effectiveness of Hive, an AI coding tool, without impacting its current functionality or auto-merge processes. The goal is to collect data on Hive's performance through new instrumentation and shadow mode experiments, providing measurable evidence for an upcoming paper on AI coding issues.

The release notes generation tool is failing because it incorrectly identifies PR numbers from commit messages that link to other PRs. This leads to errors when the tool cannot find the referenced PR in its list. The goal is to modify the tool to ignore such PR links within commit messages.

This issue proposes adding a "direct-commit" mode to a scanner, allowing direct commits to the repository under specific policy conditions (ACMM/tool-approval). It also introduces a "cadence" setting for publishing synthesized "bead" data, either per run or as a daily batch. The goal is to streamline publishing while maintaining auditability and policy adherence.

This issue proposes replacing Prow's `dco-signoff` labels with hive-owned `dco` (green) and `no-dco` (red) labels. The change aims to improve the visibility and handling of DCO (Developer Certificate of Origin) compliance, including immediate staging to `stage/072-returned-to-contributor` for missing DCOs and providing specific rebase instructions.

This issue proposes to improve the clarity of GitHub PR statuses by renaming `stage/090-needs-human` to `stage/090-ready-for-human-merge` and introducing a new `stage/085-needs-decision` label. The goal is to consolidate the decision-making process for maintainers, making it immediately clear whether a PR is ready for a human merge or requires further human intervention.

This issue describes several UI bugs on the Projects view. The "Needs human" filtered list is not correctly hiding the repo card grid, the "Swarm" header is misplaced, and the filtered list itself is not collapsible. The proposed fix involves CSS adjustments, moving elements, and adding a persistent collapse mechanism.

A warning related to a legacy `snapshot_url` configuration is repeatedly logged every time the `Validate()` function is called. This occurs not only during initial configuration loading but also during dashboard hub saves and config reloads, generating unnecessary noise. The proposed solution is to implement a mechanism that ensures the warning is displayed only once per process.