IPMC MCP
IPMC MCP
一个用于 Apache Incubator PMC 监督视图的小型无依赖 MCP 服务器。
它整合了:
来自
PodlingsMCP的孵化项目生命周期数据来自
apache-health的社区和报告信号
它提供了一系列针对孵化器层面的专业工具,以帮助 IPMC:
识别需要关注的孵化项目
评估毕业准备情况
生成孵化项目简报
标记需要导师介入的情况
总结各孵化项目的社区健康模式
要求
Python 3.12+
Related MCP server: Podlings MCP
运行
python3 server.py该服务器使用 stdio,因此旨在由 MCP 客户端启动。
安装
python3 -m pip install -e .对于开发工具:
python3 -m pip install -e .[dev]MCP 客户端配置示例
{
"mcpServers": {
"ipmc": {
"command": "python3",
"args": [
"/Users/justinmclean/IncubatorMCP/server.py",
"--podlings-mcp-repo",
"/Users/justinmclean/PodlingsMCP",
"--health-mcp-repo",
"/Users/justinmclean/HealthMCP",
"--health-source",
"/Users/justinmclean/incubator/tools/health/reports"
]
}
}
}默认运行时假设同级的源 MCP 仓库和健康报告位于:
/Users/justinmclean/PodlingsMCP/Users/justinmclean/HealthMCPreports(除非设置了--health-source)
工具调用也可以通过 podlings_source 和 health_source 覆盖源数据路径。
通过启动参数配置源 MCP 位置和健康报告目录:
--podlings-mcp-repo:PodlingsMCP检出路径--health-mcp-repo:HealthMCP检出路径--health-source: 默认的 apache-health 报告目录
测试
python3 -m unittest discover -s tests -v覆盖率
python3 -m coverage run -m unittest discover -s tests
python3 -m coverage report -m覆盖率范围限定在本地 ipmc 包内,因此导入的同级 MCP 不会稀释报告。
架构
有关模块布局、运行时流程和测试结构,请参阅 docs/architecture.md。
工具
ipmc_watchlist
根据生命周期和健康信号的综合情况,返回最需要 IPMC 关注的孵化项目。
参数:
podlings_source:podlings.xml的可选 URL 或本地文件路径health_source: apache-health markdown 报告的可选报告目录as_of_date: 用于时间敏感视图的可选YYYY-MM-DD日期limit: 可选的最大结果数severity_at_least: 可选的最低严重性过滤器include_reasons: 可选的原因过滤器列表
graduation_readiness
评估孵化项目是否准备好、接近准备好或尚未准备好毕业。
参数:
podling: 必需的孵化项目名称podlings_sourcehealth_sourceas_of_dateinclude_evidence: 可选布尔值,默认为 truestrict_mode: 可选布尔值
podling_brief
返回针对单个孵化项目的 IPMC 导向简报。
参数:
podling: 必需的孵化项目名称podlings_sourcehealth_sourceas_of_datefocus: 可选的领域列表brief_format: 可选的summary(摘要)或detailed(详细)
mentoring_attention_needed
返回似乎需要导师介入的孵化项目。
参数:
podlings_sourcehealth_sourceas_of_datelimit: 可选的最大结果数urgency_at_least: 可选的最低紧迫性过滤器include_causes: 可选的原因过滤器列表
community_health_summary
返回各孵化项目社区健康模式的 IPMC 层面摘要。
参数:
podlings_sourcehealth_sourceas_of_datescope: 可选all_podlings、active_podlings或reporting_podlingsgroup_by: 可选none、risk_band、mentor_load或age_bandinclude_examples: 可选布尔值
默认值
省略时,
podlings_source默认为 ASFpodlings.xmlURL。省略时,
health_source使用--health-source,如果未设置该启动参数,则使用reports。监督视图默认关注当前孵化项目。
健康分析优先使用以下顺序中最新的可用窗口:
3m、6m、12m、to-date。
观点模型
该服务器将源事实与派生观点分开。风险和准备情况视图源自:
导师覆盖率
报告可靠性
导师签署信号
社区活跃度
发布可见性
孵化时长
参与广度
所得出的输出旨在支持 IPMC 的判断,而非取代它。
Available Tools
21 toolscommunity_health_summaryC
Return an IPMC-level summary of community health across podlings, including strong patterns, risks, and mentoring-capacity signals.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
configure_sourcesA
Set or inspect process-level source defaults so later IPMC tool calls can omit source override arguments.
Inputs: podlings_source, health_source, report_source, mail_source, +3 more.
cross_source_mismatchesB
Return concrete mismatches between cached report narrative and current health or release evidence.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
current_podlings_overviewB
Return a factual overview of current Incubator podlings from podlings.xml lifecycle metadata.
Inputs: podlings_source, as_of_date, limit, include_descriptions.
graduation_readinessB
Assess whether a podling appears ready, near ready, or not yet ready for graduation.
Inputs: podling (required), podlings_source, health_source, report_source, +5 more.
ipmc_watchlistA
Return podlings that most need IPMC attention based on combined lifecycle and health signals.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
mentoring_attention_neededA
Return podlings where mentoring intervention appears necessary, with urgency and likely causes.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
podling_briefB
Return an IPMC-oriented briefing for a single podling, including status, trajectory, and attention areas.
Inputs: podling (required), podlings_source, health_source, report_source, +5 more.
recent_changesC
Return per-podling recent deltas the IPMC should scan, excluding unchanged or static fields.
Inputs: podlings_source, health_source, report_source, mail_source, +4 more.
refresh_mail_cacheC
Refresh cached general@incubator.apache.org message summaries used by IPMC mail evidence tools.
Inputs: mail_source, mail_api_base, mail_timespan, query, +1 more.
refresh_report_cacheB
Refresh cached ASF Incubator report data used by IPMC report-narrative tools.
Inputs: report_source, years, full_history, limit, +2 more.
release_artifact_evidenceB
Return ReleaseMCP artifact, sidecar, cadence, Incubator naming evidence, and optional release-page and GitHub/Docker Hub/PyPI/Maven distribution hints for one podling.
Inputs: podling (required), release_dist_base, release_archive_base, release_max_depth, +6 more.
release_visibilityB
Return release-governance visibility concerns, including no releases and activity/release mismatches.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
release_vote_evidenceB
Return MailMCP release vote/result thread evidence for one podling alongside IPMC release visibility signals.
Inputs: podling (required), podlings_source, health_source, report_source, +5 more.
reporting_cohortB
Return current reporting podlings grouped into non-ranked IPMC review buckets: reporting issues, release visibility issues, recent significant changes, and no obvious concerns.
Inputs: podlings_source, health_source, report_source, mail_source, +3 more.
reporting_gapsB
Return podlings with Incubator reporting compliance gaps, excluding activity analysis.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
reporting_reliabilityB
Return objective reporting reliability patterns over time, separating one-off late reporting from repeated late or missing reporting.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
reporting_scheduleB
Return expected Incubator reporting cadence, due-this-month status, and next expected reporting period from PodlingsMCP.
Inputs: podlings_source, podling, as_of_date, report_month, +2 more.
report_narrative_signalsB
Return report-derived narrative signals such as latest reported issues, recurring issues, low observed mentor sign-off, and release visibility mismatches.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
significant_changesC
Return a structured factual subset of recent changes: no 12-month releases, large activity shifts, and newly visible reporting or release transitions.
Inputs: podlings_source, health_source, report_source, mail_source, +5 more.
stalled_podlingsB
Return podlings that match the strict low-delivery, no-release stalled definition.
Inputs: podlings_source, health_source, report_source, mail_source, +3 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 21 tools
Each tool has a clearly defined and distinct purpose, focusing on specific aspects of IPMC oversight such as community health, podling status, releases, reporting, or mentoring. There is no ambiguity between tools.
All tool names follow a consistent pattern of descriptive phrases using underscores (e.g., community_health_summary, graduation_readiness). No mixing of camelCase or other conventions is observed.
With 21 tools, the server covers a broad domain of IPMC responsibilities. While the count is relatively high, each tool addresses a necessary function, and the scope justifies the number.
The tool set provides comprehensive coverage of IPMC information needs, including community health, podling lifecycle, releases, reporting, and mentoring. Minor gaps may exist for direct actions, but the analysis surface is thorough.
Maintenance
Related MCP Connectors
MCP server for VC pitch-deck scoring, thesis-fit matching, and deal-flow management.
MCP server for CertScore.ai public website risk-signal workflows.
Guarded MCP server for agent-readable business truth, provenance, readiness, and discovery.
One MCP server for all your MCP servers: code mode, per-tool permissions, approvals and audit.
Related MCP Servers
- FlicenseAqualityDmaintenanceA deterministic, network-free MCP server for validating repository release hygiene and version alignment in local projects. It enables automated repository health checks and generates standardized release checklists based on project state.1-
- AlicenseBqualityDmaintenanceAn MCP server for accessing and analyzing Apache Software Foundation Incubator podling data from podlings.xml. It provides tools to query podling metadata, generate statistics, and analyze incubation trends over time.22MIT
- AlicenseAqualityDmaintenanceAn MCP server for querying and analyzing Apache Incubator podling health reports. It provides tools to list, search, and compare podling metrics across different time windows from Apache's health report Markdown files.9MIT
- AlicenseNot gradedqualityDmaintenanceMCP server for comprehensive PyPI package intelligence, providing tools for dependency analysis, security scanning, health scoring, license compliance, and trend tracking.MIT