メニュー
Bro's PR/MR comment triage — resolve the PR, fetch the comment threads, judge which are task-worthy, and dispatch SWE per ratified group. Loaded by bro when /monitor surfaces PR/MR comments.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Bro's response when something fails — an MCP tool returns is_error=true (halt + surface), the trajectory-server is unreachable (degraded sqlite3 readonly fallback), or the world model is unavailable (retry, identify the lock holder, reconnect the binding). Loaded reactively on the first failure of a session.
Bro's code-touching flow — verify world model, triage the requirement (reuse vs build), propose branch, write spec, dispatch SWE, verify on return, atomic-close. Loaded on the first code-touching ask of a session.
When bro hits a wall — a task leans on a capability the project plainly lacks — and an existing published skill, MCP toolkit, or plugin would close the gap better than hand-rolled code. Bro names the gap, calls cheatcode_search for ranked candidates, judges which best fits this task and codebase, and recommends it for the Human to approve. Loaded when grabbing an external cheatcode beats grinding the capability out from scratch.
Bro's push-gate orchestration — reaping unsigned commits, spawning pr-reviewer per task, and the all-pass push + PR-create + post-merge cleanup path. Loaded by bro when the push hook blocks or the Human asks for review-before-push.
PR-reviewer's diff-level push-gate review — check this task's diff against its spec, then write the validation_record verdict that gates the push. Loaded by pr-reviewer when reviewing a committed task. Self-contained.
Generate a new project-local skill at .claude/skills/<name>/SKILL.md and attach it to existing agents. Loads when the user asks to capture a repeatable behavior — e.g. "create a skill that codifies <our convention>", "teach swe to also <run mypy / use black / etc.>", "make a skill for <reviewing PRs / writing changelogs / etc.>", "we need a checklist when <X happens>". Extends the consuming agent's `skills:` frontmatter array only; agent body stays intact. Always Human-approved.
| name | tmb_comment-triage |
| description | Bro's PR/MR comment triage — resolve the PR, fetch the comment threads, judge which are task-worthy, and dispatch SWE per ratified group. Loaded by bro when /monitor surfaces PR/MR comments. |
| allowed-tools | Task, AskUserQuestion, mcp__plugin_tmb_trajectory-server |
pr_monitor_comments_get does the deterministic fetch + since-marker bookkeeping. This skill is the judgment around what's task-worthy.
If the Human named a PR number, use it. Otherwise:
gh pr view --json numberglab mr list --source-branch <branch> --jsonEmpty result → ask the Human which PR/MR number to monitor (free-text answer).
Call pr_monitor_comments_get with the PR number.
Carrier: resolve the linked issue with branch_report_md for the current branch rather than reading the DB directly. If unresolved, ask the Human which issue the PR is linked to (free-text answer).
Skip as informational when:
Treat as task-worthy when the comment names a concrete change request (should be, please change, consider X over Y, ends with ?), or contains a code suggestion fence.
Group task-worthy comments by file or shared concept; one task per group. Flag tasks that touch DB schema files (e.g. schema/*.sql), public API surfaces, or config directories (e.g. agents/, skills/, plugin manifest files) as (arch-impact).
Offer the ratified groups as a multi-select — one option per group, titled by the task it would become (suffix arch-impact ones) — and let the Human pick a subset.
For each ratified group: task_create_batch(...), spawn SWE, and if arch-impact, invoke scan_run(source='bro_auto_post_change') after SWE returns to refresh the world model.