mcp-github-project-manager
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GITHUB_REPO | Yes | GitHub repository name | |
| AI_PRD_MODEL | No | AI model for PRD generation (default: claude-3-5-sonnet-20241022) | claude-3-5-sonnet-20241022 |
| GITHUB_OWNER | Yes | GitHub repository owner (username or organization) | |
| GITHUB_TOKEN | Yes | GitHub personal access token with 'repo', 'project', and 'write:org' permissions | |
| AI_MAIN_MODEL | No | Main AI model to use (default: claude-3-5-sonnet-20241022) | claude-3-5-sonnet-20241022 |
| GOOGLE_API_KEY | No | Google AI key for Gemini models | |
| OPENAI_API_KEY | No | OpenAI API key for GPT models | |
| AI_FALLBACK_MODEL | No | AI model for fallback support (default: gpt-4o) | gpt-4o |
| AI_RESEARCH_MODEL | No | AI model for research tasks (default: perplexity-llama-3.1-sonar-large-128k-online) | perplexity-llama-3.1-sonar-large-128k-online |
| ANTHROPIC_API_KEY | No | Anthropic API key for Claude models | |
| MAX_SUBTASK_DEPTH | No | Maximum depth for task expansion (default: 3) | 3 |
| MAX_TASKS_PER_PRD | No | Maximum number of tasks to generate per PRD (default: 50) | 50 |
| PERPLEXITY_API_KEY | No | Perplexity API key for research and analysis tasks | |
| AUTO_CREATE_LIFECYCLE | No | Automatically create task lifecycles (default: true) | true |
| AUTO_EFFORT_ESTIMATION | No | Automatically estimate task effort (default: true) | true |
| ENHANCED_CONTEXT_LEVEL | No | Level of enhanced context: minimal, standard, or full (default: standard) | standard |
| AUTO_GENERATE_USE_CASES | No | Automatically generate use cases for features (default: true) | true |
| AUTO_CREATE_TRACEABILITY | No | Automatically create requirements traceability matrix (default: true) | true |
| ENHANCED_TASK_GENERATION | No | Enable enhanced task context generation (default: true) | true |
| INCLUDE_BUSINESS_CONTEXT | No | Include business context in task generation (default: false) | false |
| AUTO_DEPENDENCY_DETECTION | No | Automatically detect task dependencies (default: true) | true |
| INCLUDE_TECHNICAL_CONTEXT | No | Include technical context in task generation (default: false) | false |
| DEFAULT_COMPLEXITY_THRESHOLD | No | Default threshold for task complexity analysis (default: 7) | 7 |
| INCLUDE_IMPLEMENTATION_GUIDANCE | No | Include implementation guidance in task generation (default: false) | false |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": true
} |
| resources | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| manage_projectC | Manage GitHub Projects (v2): create, list, get, update, delete projects; manage readme, fields, views, items; handle templates and link to repos/teams. Use the |
| manage_issuesB | Manage GitHub Issues: create, list, get, update issues; manage comments and drafts; search with advanced filters; manage sub-issues. Use the |
| manage_prsC | Manage Pull Requests: create, get, list, update, merge PRs; list and create reviews. Use the |
| manage_milestonesC | Manage Milestones: create, list, update, delete milestones; get metrics; find overdue and upcoming milestones. Use the |
| manage_sprintsB | Manage Sprints: create, list, update, delete sprints (delete requires confirm: true); get current sprint; add/remove issues; get metrics and plan. Use the |
| manage_labelsB | Manage repository Labels: create and list labels. Use the |
| manage_automationB | Manage Project Automations: list, get and delete the automation workflows of a GitHub project. Use the |
| manage_iterationsC | Manage Project Iterations: get configuration, current iteration, items; find by date; assign items. Use the |
| manage_eventsC | Manage Events: subscribe to project events, get recent events, replay events. Use the |
| manage_status_updatesC | Manage project Status Updates: create, list, and get status updates. Use the |
| ai_generateC | AI-powered generation: generate/enhance/parse PRDs, add features, get next tasks, analyze complexity, expand tasks, create traceability matrices. Use the |
| ai_analyzeC | AI-powered analysis: enrich issues (single/bulk), triage issues, schedule triaging, suggest labels, detect duplicates, find related issues. Use the |
| ai_planC | AI-powered planning: calculate capacity, prioritize backlog, assess risk, suggest sprint composition, generate roadmaps and visualizations. Use the |
| agent_workC | Agent work operations: register agents, checkout/release/complete tasks, send heartbeats, submit/check review state, validate work products, and get handoff context. Use the |
| agent_manageC | Agent management: list/deregister agents, activity, work products, budgets (warningFraction), stale reclaim, smart assignment, workload rebalancing, convergence, and registry cleanup. Use the |
| systemC | System operations: health check and project field setup. Use the |
| discover_toolsA | Discover available compound tools, their actions, and parameters. Always available regardless of group filter. |
| manage_branchesC | Manage GitHub Branch Protection: get, update, and delete a branch's protection rules; list repository branches. Use the |
| manage_releasesB | Manage GitHub Releases: create, list, get, update, delete releases; get the latest release. Use the |
| manage_workflowsA | Manage GitHub Actions Workflows: list workflows, trigger dispatch events, inspect run status and logs, list runs, and cancel runs. Use the |
| analyticsB | Read-only project analytics: sprint velocity and burndown, cycle time, throughput, and a project summary. Use the |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| review-prd | Ask the AI to review a Product Requirements Document for clarity, completeness, and quality. |
| explain-task-complexity | Ask the AI to analyze and explain the complexity of a task. |
| suggest-issue-labels | Ask the AI to suggest appropriate labels for a GitHub issue. |
| analyze-sprint-risk | Ask the AI to assess the risk profile of an upcoming sprint. |
| generate-release-notes | Ask the AI to draft user-facing release notes from milestone information. |
| triage-issue | Ask the AI to perform first-pass triage on a new GitHub issue. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| config | Target repository, GitHub auth mode, and AI provider availability. |
| projects | Active GitHub Projects (v2) for the configured repository. |
| current-sprint | The currently active sprint, or null when none is active. |
| milestones | Open milestones for the configured repository. |
| agents | Agents registered in the orchestration swarm registry. |
| tools | Catalog of registered MCP tools, keyed by name. |
TDQS
Scored across 21 tools
Most tools are cleanly separated by GitHub resource or domain: issues, PRs, releases, workflows, milestones, sprints, labels, and analytics each have a distinct target. However, there is meaningful overlap between agent_manage and agent_work, and between ai_plan, ai_generate, ai_analyze, and analytics, which could cause misselection for planning/analysis tasks.
The naming is consistently snake_case and mostly follows predictable prefixes such as manage_*, ai_*, and agent_*. Minor deviations exist: some names are noun-only (analytics, system), and singular/plural patterns vary (manage_project vs manage_issues, manage_prs abbreviation).
At 21 tools, the server is on the heavy side for an MCP surface, though the use of compound action-based tools mitigates some complexity. The broad project-management scope justifies many categories, but the count is still borderline and could overwhelm agents choosing among similar AI/agent tools.
The surface covers a broad GitHub project-management lifecycle: projects, issues, PRs, milestones, sprints, iterations, releases, workflows, branch protection, analytics, AI assistance, and agent coordination. Some minor gaps remain, such as limited label lifecycle operations and incomplete comment/review management, but core workflows are well covered.