mattermost-ai-marketplace

Add an MCP (Model Context Protocol) server to a Mattermost plugin so the Agents plugin can call its tools. Use when implementing cross-plugin MCP, exposing AI tools from a Mattermost plugin to the Agents plugin, or wiring up the `pluginmcp` helper from mattermost-plugin-agents.

Emits the skimmable two-layer skim block at the TOP of every phase artifact — TL;DR, phase + complexity tier, what-changed-since-last-version, decisions locked (and which flipped from a Recommended default), open [VERIFY WITH PM] items, and a one-line reading guide. Goal — a reader answers what / phase / what-changed / what's-open in under 60 seconds without reading the body. Composable; phase agents invoke it as their first output step.

Cherry-pick a single merged commit onto one release-X.Y branch, correctly resolving conflicts, running lint in a separate commit, pushing, and opening a cherry-pick PR via the create_pr_tool MCP following the PR template. Self-contained (carries its own conflict-resolution logic). Use inside a per-branch subagent when backporting a merged PR to a release branch. Has side effects (push + open PR).

Reusable contract for surfacing multiple-choice clarifying questions (with a recommended option) at phase intake and whenever in-phase ambiguity exceeds threshold. Defines question shape, threshold scoring, capture format, and bypass. Phase-specific question banks live in the calling agent. Surface-and-pause is mandatory; the agent never self-resolves Recommended options.

Audit and improve CLAUDE.md files in repositories. Use when user asks to check, audit, update, improve, or fix CLAUDE.md files. Scans for all CLAUDE.md files, evaluates quality against templates, outputs quality report, then makes targeted updates. Also use when the user mentions "CLAUDE.md maintenance" or "project memory optimization".

Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality. Focuses on recently modified code unless instructed otherwise.

Resolve git merge conflicts by analyzing change intent across branches. Use when the user asks to resolve merge conflicts, sync a branch, merge branches, or when a git merge produces conflicts.

Create a new Mattermost plugin from the starter template in the current directory. Use when creating a new plugin from scratch, scaffolding a Mattermost plugin, or bootstrapping a plugin project.

Use the `cursorctl` CLI (Cursor SDK Bridge) to spawn, send, detach, follow up, inspect, and collect Cursor cloud or local agents. Trigger when delegating work to Cursor cloud agents (`bc-…` IDs), running a local Cursor agent, fire-and-forget with `--detach`, polling `run get`/`wait`/`watch`, collecting artifacts or conversations, managing `cursorctl auth` profiles, or troubleshooting the SDK Bridge. Run cursorctl via Bash. Do not call the SDK Bridge HTTP API directly.

Categorizes raw stakeholder feedback into actionable priority tiers with suggested fixes for gate decisions

Translate Figma nodes into production-ready code with 1:1 visual fidelity using the Figma MCP workflow (design context, screenshots, assets, and project-convention translation). Trigger when the user provides Figma URLs or node IDs, or asks to implement designs or components that must match Figma specs. Requires a working Figma MCP server connection.

Investigate production issues, query logs and metrics, and explore dashboards on the Mattermost Grafana instance at grafana.internal.mattermost.com.

Autonomous issue resolution agent that investigates, plans, and fixes issues using sub-agents and browser-based validation. Use when the user provides one or more issues to resolve autonomously, wants hands-off debugging and fixing, or needs an orchestrated investigate-plan-execute-validate workflow.

Backfill realistic test data into a Mattermost server using the Mattermost MCP tools. Creates users, teams, channels, and natural conversations. Use when the user asks to populate a Mattermost instance, create test data, set up a demo environment, seed conversations, or backfill a Mattermost server. Also provides guidance on reading, searching, and interacting with Mattermost via MCP tools.

Given a Mattermost priority/severity level and a plugin repository, resolve the target plugin release-X.Y branches by cross-referencing the platform release policy with the plugin versions declared in each platform Makefile, plus any plugin release branches not yet wired into the platform. Returns only branches that exist on the plugin's origin.

Discover and run all pre-commit code quality checks (linters, type checks, i18n extraction) in a monorepo, auto-fix failures, and report results. Use when the user asks to run pre-commit checks, validate code before committing, or ensure CI checks pass locally.

Normalizes an arbitrary Mattermost PRFAQ markdown file (T3-aware) into a structured, format-independent PRFAQ extract — feature intent, tier, actors, named surfaces, capabilities/actions, behavioral rules, verbatim UI copy, scope, open design questions, dependencies — reconciling in-document contradictions and scrubbing customer names. Feeds scene-mapper and the prfaq-analyst agent.

Target-profile-driven component composition. Enumerates the active playground's real component library at runtime, reads each component's props before use, and fills scaffolded scene files with typed JSX, defense-realistic demo data, and any verbatim PRFAQ copy — flagging gaps instead of inventing phantom imports. Swappable via ${CLAUDE_PLUGIN_ROOT}/config/prototype-targets.json.

Target-profile-driven scaffolding for a multi-scene, multi-option prototype into the active playground defined in ${CLAUDE_PLUGIN_ROOT}/config/prototype-targets.json. Reads the target's real conventions at runtime (folder layout, manifest shape, scene API), checks for collisions/reuse, and creates the orchestrator + scene files + manifest entry. Swappable — point at a newer playground by changing the target profile, not this skill.

Turns a PRFAQ extract into a ranked screen/scene inventory, a set of genuine option-divergence axes for multi-option prototyping, an assumptions ledger, and a rough scene→component feasibility map against the active target playground. Feeds the prfaq-analyst clarification round and the prototype-builder.