dougborg

Close a GitHub issue with discipline — resolved, superseded, or duplicate. Pre-flights body + comments, migrates substance before destroying context, cross-links both directions.

File a new GitHub issue with duplicate search, scope decision, label discovery, and preview before posting. Prevents fragmented or silently-filed issues.

Restructure the topology of GitHub issues — split one into many focused replacements, or merge multiple into one keeper. Migrates substance and cross-links before closing anything.

Update an existing GitHub issue — edit body, post a comment, retag, or reopen. Pre-flights body + comments; acknowledges stale framing instead of silently rewriting.

Review and address PR feedback using 6-dimensional code review

Structured code review using 6 dimensions. Works with or without the code-reviewer agent.

Create conventional commits with quality gates and validation

Write scannable, progressive-discovery documentation using layered contract pattern

Write comprehensive tests using project conventions: AAA pattern, edge-case checklist, httpx.MockTransport, conftest fixtures, parametrize where it reads better than copy-paste.

Audit Prefab UI card builders (`build_*_ui` in `katana_mcp_server/src/katana_mcp/tools/prefab_ui.py`) for user-centric content: real names over IDs, no redundant text dumps, named helpers over ad-hoc rendering. Reports findings; does not modify code. Use when reviewing card UX or after touching `prefab_ui.py`.

Groom the Rolling Backlog (project #5) by reading current state and proposing diffs — not re-deriving the queue from scratch. Surfaces drift between issue state and project status, missing classification, and stale priorities. Asks per-category confirmation before mutating the board.

Quick pre-flight validation of staged and unstaged changes before committing. Lighter than the verifier/code-reviewer agents — focuses on catching common mistakes in the current changeset.

Scan editable code for tech debt, anti-patterns, and improvement opportunities across 5 categories. Reports findings; the code-modernizer agent applies fixes.

Open a PR for the current feature branch — self-review the diff, organize commits, push, create the PR, wait for CI and review, then address feedback. Use when implementation is complete and ready for review.

Address PR review comments — fetch all unresolved comments, fix the issues, validate with `uv run poe check`, commit, push, and reply to each comment on GitHub. Use when asked to handle PR review feedback.

Resolve a shipped-order discrepancy — customer received different items than Katana recorded. Pull the customer-facing record, locate the linked Katana SO/MO, decide whether a BOM-driven split is needed (when an assembled good shipped as its component parts), then apply correct_manufacturing_order + correct_sales_order with operator-judgment gates between phases. Verify against the source of truth at the end.

Open a PR for the current feature branch — validate, self-review, simplify, organize commits, push, create the PR, wait for CI and review, then address feedback. Use when implementation is complete and ready for review.

Respond to PR review comments systematically in thread context

Review and address PR feedback using 6-dimensional code review

Add a field to NULLABLE_FIELDS in scripts/regenerate_client.py and regenerate the client. Use when a TypeError appears in logs or tests because a generated model expected a non-null value but the StockTrim API returned null.

Regenerate the StockTrim OpenAPI client from the live spec, run quality checks, and commit the result. Use when the StockTrim API spec changes, a new endpoint is added, or NULLABLE_FIELDS in scripts/regenerate_client.py is updated.

Add a new MCP tool to stocktrim_mcp_server, including the tool function, service layer, and test. Use when exposing an existing helper to AI clients via the Model Context Protocol.

Run the full CLAUDE.md zero-tolerance quality gate before declaring any task done. Checks lint, type, tests (root + MCP), forbidden patterns, feature branch, and commits. Use as the final pre-PR check.

Add a new ergonomic helper method to stocktrim_public_api_client/helpers/ that wraps a generated API call. Includes test scaffolding and quality gate. Use when adding a domain operation (e.g., bulk upsert, find-by-code) that callers should reach for instead of the raw generated client.

Open a PR for the current feature branch — validate, self-review, simplify, organize commits, push, create the PR, wait for CI and review, then address feedback. Use when implementation is complete and ready for review.

Write pytest tests using AAA pattern, project naming convention, and helper-based assertions (unwrap_as, unwrap_data, unwrap_unset). Delegates to the test-writer agent.

Review the current branch for correctness, security, architecture, and project conventions before opening a PR. Delegates to the code-reviewer agent and adds StatusPro-specific checks.

Structured code review using 6 dimensions. Works with or without the code-reviewer agent.

Create conventional commits with quality gates and validation

Generate or update docstrings, ADRs, READMEs, cookbook recipes, and user guides. Knows about MCP help-resource drift and ADR numbering scheme.

Respond to PR review comments systematically in thread context

Scan editable code for tech debt and anti-patterns specific to this repo (UNSET misuse, manual status-code checks, retry wrapping, dead code). Reports findings without applying fixes.

Scaffold a new resource vertical end-to-end (helper + Pydantic domain + MCP tools + tests + docs). Use when starting a new "client.<resource>" surface such as drafts (#14), contacts (#2), messages (#4), tags/inboxes (#5). Requires the `vertical-planner` agent to produce a plan first.

Open a PR for the current feature branch — self-review the diff, organize commits, push, create the PR, wait for CI and review, then address feedback. Use when implementation is complete and ready for review.

Watch CI on multiple open PRs in parallel, surface real failures with their job-log tails, and chain into `/review-pr` when comments arrive. Use when 2+ PRs are open at once and individually polling each would burn context.

Address PR review comments — fetch all unresolved comments, fix the issues, validate with `uv run poe check`, commit, push, and reply to each comment on GitHub. Use when asked to handle PR review feedback.

Refresh the vendored Front OpenAPI spec from upstream and regenerate the Python client safely. Use when pulling upstream API changes, when a new endpoint is needed, or when debugging codegen drift. NEVER edit the vendored spec or generated files directly — patch scripts/vendor_spec.py instead.

Add a new GAS global or method mock to the `installGasGlobals()` factory in `packages/gas-test-utils`. Use when a test in any package needs a GAS API that is not yet stubbed (a missing service, or a missing method on an existing service).

Scaffold a new `packages/gas-*` package in this monorepo with the correct structure (src/, package.json, tsconfig.json, README.md, CHANGELOG.md, LICENSE) following existing conventions. Use when adding a new GAS-tooling package to publish under the `@dougborg/*` scope.