Committer readiness and onboarding
Scope —
organization: ASF· 🪶 ASF-specific. This family encodes Apache Software Foundation processes (the contributor-to-committer path) and assumes an ASF adopter profile by default. Non-ASF projects can still adopt it through the adapter/config layer, but it carries ASF assumptions the generic families do not.
This plugin contains nine maintainer-facing skills for contributor activity,
nomination evidence, identity mapping, and committer onboarding. Newcomer
welcomes, good-first-issue authoring, and backlog curation live in the separate
magpie-mentoring plugin.
Why a framework skill family? The contributor-to-committer path is one of the highest-leverage levers an open-source project has for long-term health — lowering onboarding friction and shortening the time from first PR to committer status keeps the contributor pipeline healthy. These skills were designed independently but cover a contiguous path; grouping them makes the adopter configuration and the evaluation story coherent.
Surface information, never rank
The skills that look at people — activity-sweep, calibrate, candidate-screen, contributor-to-committer and nomination — only surface information.
The decision about who is nominated, and when, is always made by PMC members.
- Never a ranking. No skill orders people by any measure, scores them against each other, or picks a “top” few. Every list of people is in alphabetical order of GitHub handle, compared case-insensitively.
- Never a verdict.
No skill says or suggests whether someone is ready, close, or not ready to be nominated.
Counts are shown next to the project’s reference levels as plain numbers, without a status, a band or a recommendation.
candidate-screengoes further: it uses the reference levels only to pre-filter out the long tail, and shows everyone else’s data with no reference levels at all. - Deliberately more, not fewer.
A list of likely candidates includes more people than the PMC would consider, on purpose, so that nobody is overlooked.
calibratesets its reference levels well below what the project has actually elected — by default three quarters of it (calibration_relaxation,0.75) — to make that true. Appearing on a list means only that someone’s details are worth a look. - A report is a list and a summary. A report lists the people it covers, alphabetically, each linked to their details, followed by a paragraph or two summarising what was found. The summary describes the findings; it does not judge or recommend anyone.
Every brief and report these skills produce says this at the top.
Install & first runs
Install & first runs has the install command for this family, the files to configure before the first run, and the skills to try first.
Why the configuration is personal
This family is personal-recommended and not adoptable by default.
Install it for yourself — user scope, or install-only in this repository — on the machine of each maintainer who runs these skills.
/magpie-setup:setup adopt leaves it out of the committed floor unless a maintainer explicitly insists and accepts the risk below, and records that acceptance in .apache-magpie.lock.
The reason is gaming. Committer thresholds, nomination criteria, calibration floors, sentiment caps and the identity map are the project’s private judgement about people. Committed to the repository, they become a public checklist: a contributor can point at a threshold and demand a nomination, and every later edit to it becomes a negotiation rather than a decision.
So the five files this family adds — committer-onboarding-config.md, committer-readiness.md, contributor-identities.md, contributor-nomination-config.md and contributor-sentiment-config.md — always live in your personal layer, even in a project that has adopted Magpie and even when the family itself is in the floor:
<git-common-dir>/apache-magpie/ when Magpie is only installed, .apache-magpie-local/ when the project has adopted it.
Never commit them to .apache-magpie-overrides/;
adopt does not promote or scaffold them there, and the skill validator warns when it finds one.
contributor-calibrate writes its floors to the personal layer and never offers the committed one.
Maintainers who want the same numbers share them the way they share the deliberation itself, on the private list.
Stage coverage
| Stage | Skill | What it does |
|---|---|---|
| Activity tracking | contributor-activity-sweep |
Produces a read-only GitHub activity card (PRs authored, code reviews, issues, comments) over a configurable window. |
| Calibration | contributor-calibrate |
Derives committer and PMC reference levels from the project’s own past nomination decisions on the private list, deliberately relaxed well below what the project elected; proposes a config diff holding numbers only, with nothing about any nominee leaving the session. |
| Candidate screening | contributor-candidate-screen |
Pre-filters out the long tail of recent contributors, then shows the activity data of everyone else — deliberately more people than the PMC would consider, never scored or compared with thresholds — and writes a report to a repository the GitHub API reports as private, after the maintainer has read it: a Headline split by target (committer / PMC) with one phrase per person, each full section linked to its detail, and a paragraph or two summarising the findings. Never a ranking, never a verdict. |
| Activity vs. reference | contributor-to-committer |
Shows a contributor’s activity next to the project’s committer or PMC reference levels, as plain numbers with the difference, and a short factual summary. No status, band, or readiness verdict. Read-only; never opens a nomination thread. |
| Identity mapping | contributor-identity-map |
Maps any contributor’s GitHub handle to their Slack, Discord, Matrix, mailing-list, and social-media handles; infers from the sources the session can reach and records only the mappings a maintainer confirms. Used by committer-onboarding and contributor-nomination. |
| Nomination brief | contributor-nomination |
Assembles evidence for a committer or PMC discussion: activity breadth, consistency, vendor-neutrality context, and factual narrative prose. Never rates the contributor or says whether they are ready. Read-only; never posts to any list. |
| Sentiment analysis | contributor-sentiment |
Analyse contributor sentiment signals (issue tone, PR abandonment, response-time frustration) to surface early-warning indicators of contributor disengagement. Read-only. |
| Onboarding concierge | onboarding-concierge |
Interactive first-session guide for new contributors: walks through repo setup, points to good first issues, introduces project conventions and communication channels. |
| Post-vote onboarding | committer-onboarding |
Walks the nominator through ICLA check, account provisioning, permissions grant, and the welcome announcement for committer and PMC promotions at ASF TLPs and podlings. |
Every stage is read-only on governance artefacts or propose-before-post: no skill modifies a roster, posts an announcement, or files an issue without explicit maintainer confirmation.
Related newcomer mentoring skills are installed by magpie-mentoring, not
this plugin: mentoring-welcome,
good-first-issue-author,
and good-first-issue-sweep.
See the mentoring overview for their purpose and
installation instructions.
Skills
| Skill | Mode | Status |
|---|---|---|
contributor-sentiment |
Triage | experimental |
onboarding-concierge |
Mentoring | experimental |
contributor-activity-sweep |
Triage | experimental |
contributor-identity-map |
Triage | experimental |
contributor-to-committer |
Mentoring | experimental |
contributor-nomination |
Triage | experimental |
contributor-calibrate |
Triage | experimental |
contributor-candidate-screen |
Triage | experimental |
committer-onboarding |
Triage | experimental |
All nine skills installed by this plugin are experimental.
No adopter has run the full contributor-to-committer path under evaluation conditions yet.
Family boundary
This family sits alongside two related skill families with separate scopes:
docs/mentoring/README.mddescribes the separatemagpie-mentoringplugin for newcomer welcomes, good-first-issue authoring, and backlog curation. Its skills are not installed bymagpie-contributor-growth.docs/issue-management/README.md— general-issue triage and fix workflow. The contributor-growth family reads GitHub activity data about contributors, not issue content; the two families use different query surfaces and produce different artefacts (activity cards and nomination briefs vs. issue disposition proposals).
Skills in this family propose every state-changing action for human
sign-off. committer-onboarding emits paste-ready command recipes the
nominator executes as themselves; no skill submits an ICLA form, invites
an account, or modifies repository permissions without the nominator’s
direct action.
Status
Experimental. All nine installed skills are on main with eval suites. No adopter has run the full contributor-to-committer path end-to-end under evaluation conditions.
PMC-member promotion is covered: contributor-nomination takes
target:pmc, and committer-onboarding handles the committer-to-pmc and
direct-to-pmc scenarios.
Not built yet: emeritus / inactive-committer handling and contributor offboarding. These involve project-level governance decisions (roster policy, access removal, farewell communication norms), so the proposed spec requires every policy knob to be configured before anything runs, and keeps the read-only inactivity report and the emeritus move strictly apart.
Cross-references
docs/modes.md§ Triage — mode taxonomy the three Agentic Triage-mode family skills declare against.docs/modes.md§ Mentoring — mode taxonomy the three Agentic Mentoring-mode family skills declare against.docs/mentoring/README.md— the Agentic Mentoring mode family overview, which cross-referencesmentoring-welcome,good-first-issue-author, andgood-first-issue-sweep.projects/_template/README.md— adopter scaffold index, with a purpose line for every file the table under Before the first run links.- Project rules — the override mechanism every skill in this family supports.