pmi-project-management
Skill0 starsChanged 6 days agoMITView on GitHub
Life-cycle frame and shared templates for the PM kit. Use when the task spans several PM artifacts, asks where something fits in the PMI Process Groups or Knowledge Areas, asks which artifact or command to use next, or needs an artifact with no dedicated skill (charter/PID, RACI, requirements, communications plan, meeting agenda/minutes, lessons learned, closure, issue log, decision log). Risk, stakeholder, status, change-control, and agile work have their own skills — prefer those.
In this repository6 other files it gives its agents.
- .claude/CLAUDE.mdCLAUDE.md
- agile-deliverySkill
- change-controlSkill
- risk-managementSkill
- stakeholder-managementSkill
- status-reportingSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
If this project has no project-context.md and no .claude/CLAUDE.md, the kit was installed as a plugin but the project is not set up yet: …
- Distinctive: line 79
Follow the data governance note in the project context.
- Very specific: line 81
Read the existing file in docs/ before writing.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
PMI Project Management
A methodology pack for running projects with Claude as a drafting and reasoning partner. Claude is not a system of record: it never reads live Jira/budgets/Project files and never invents figures or dates.
1. Always establish context first
If this project has no project-context.md and no .claude/CLAUDE.md, the kit was
installed as a plugin but the project is not set up yet: suggest /project-init first.
The project context block from project-context.md is loaded with CLAUDE.md. Fields
that are blank or still <...> are unknown: ask once for what the task needs, then proceed
with clearly labelled assumptions. State the audience and output format for every
deliverable.
2. Map every task to the life cycle
Name the Process Group and Knowledge Area each artifact serves.
| Process Group | Typical artifacts |
| Initiating | Charter / PID, initial stakeholder register |
| Planning | Scope & requirements, WBS, schedule, budget, RACI, risk register, comms plan |
| Executing | Meeting agendas/minutes, status updates, stakeholder comms |
| Monitoring & Controlling | Status reports, change requests, risk reviews, issue log, decision log, variance analysis |
| Closing | Lessons learned, closure summary, benefits handover |
Knowledge Areas: Integration · Scope · Schedule · Cost · Quality · Resource · Communications · Risk · Procurement · Stakeholder.
3. Templates — the single source for structure
Each artifact's structure lives in
${CLAUDE_PLUGIN_ROOT}/.claude/skills/pmi-project-management/templates/. Read the matching
template and tailor it to the project context; do not restate its structure elsewhere.
Each template ends with a ready-to-send checklist in an HTML comment: 4–7 pass/fail
items checkable from the document alone. Commands run it after drafting, and
/project-check and the reviewer agents read it from the template, so a per-artifact check
is written there once. A check that compares two files belongs in /project-check.
Checklist items never demand a fixed count or forbid [TBC]; they ask that every [TBC]
is listed under the artifact's open questions.
| Template | Artifact | Command |
charter.md | Charter / PID | /charter |
stakeholder-register.md | Stakeholder analysis + engagement | /stakeholder-analysis |
risk-register.md | Risk register, scoring scale, severity bands | /risk-register |
raci.md | Responsibility assignment matrix | /raci |
requirements.md | Requirements register + review findings | /requirements-review |
comms-plan.md | Communications management plan | /comms-plan |
agenda.md | Meeting agenda | /agenda |
minutes.md | Meeting minutes | /minutes |
issue-log.md | Issue log (things that have already happened) | /issue-log |
decision-log.md | Decision log | /decision-log |
project-status.md | Current status data the PM maintains (copied to the repo root) | /import-project |
status-report.md | Period status report, default RAG definitions | /status-report, /weekly-monday |
change-request.md | Change request | /change-request |
sprint-plan.md | Iteration / sprint plan (agile) | /sprint-planning |
retro.md | Iteration retrospective (agile) | /retro |
lessons-learned.md | Lessons learned | /lessons-learned |
closure.md | Project closure summary | /closure |
4. Draft inline; use agents for a second pass
Draft artifacts in the main conversation, where the pasted data is and where the user
can answer questions. The specialist agents (risk-manager, stakeholder-analyst,
status-reporter, requirements-analyst, meeting-scribe, pm-thinking-partner) start
without the conversation, cannot ask questions, and do not write files. Use them for an
independent review of a saved artifact, and pass the file path plus anything else they
need in the brief.
5. Discipline (non-negotiable)
- Mark unknown metrics, costs, and dates
[TBC]; never fabricate. - Separate facts (from supplied data) from PM judgement (labelled).
- Flag stakeholder-sensitive and political decisions for the human PM.
- Follow the data governance note in the project context; if it is unfilled, remind the user once per session to check policy before handling confidential/client data.
- Read the existing file in
docs/before writing; update in place and show what changes. - End artifacts with an "Open questions / data I still need" section.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

