- Home
- Skills
- Business & Operations
- StreamTeem Program Manager Expert
Works with the AI Agents you already use
StreamTeem Program Manager Expert
The pack asks what outcome is due and by when, what is in and out of scope, who governs the work and who accepts a milestone, and which dependencies are already known.
$4.99
StreamTeem Program Manager Expert
Example session with this skill installed
Start a program for the Q4 data migration. Use work item id PROG-482 and the Program Delivery workflow. No outside references yet.
- Read your context and instructions
- Compiled the streamteem program manager
Work item PROG-482 is set up on the
Program Delivery workflow. Your role was already defined, so I went straight to the work item and its first phase, intake.
Where things stand
| Item | Value |
|---|---|
| Work item | PROG-482, Q4 data migration |
| Workflow | Program Delivery |
| Phase | Intake |
| Journal | program-manager/journals/work-items/PROG-482/progress.md |
| Brief (not written yet) | program-manager/work-items/PROG-482/docs/charter/brief.md |
Four questions before I write the brief
These are blocking, and "continue" does not count as an answer.
- Outcome: What outcome is due, and by when?
- Scope: What is in scope and out of scope?
- Governance: Who governs this, and who accepts a milestone?
- Dependencies: Which dependencies are already known?
Next steps
- Answer the four questions. I record each answer in the journal in the same turn, then write the brief with the answers, the unknowns, and the risks.
- After the brief comes the charter and its diagrams: milestone and dependency map, RACI, and governance path.
- I then stop until you say "I approve the charter" or "Skip charter approval". Only after that do I write the operating plan.
Connects securely to your tools. The creator never sees your data.
About this skill
Program Manager Expert makes your agent hold a cross-team effort to a date the way a program manager does: scope, governance, dependencies, RAID, and steering status, with the decisions written down.
Say "start". The pack asks what outcome is due and by when, what is in and out of scope, who governs the work and who accepts a milestone, and which dependencies are already known. It writes a brief, then a charter with its diagrams, and stops for your approval. The operating plan follows (milestones, dependencies, RAID, RACI, communications cadence, decision rights) with a second approval gate. Only then does it write the action list, create the integrated tracker, the RAID log, the steering cadence, and the stakeholder list, and work through the actions one at a time: status reports, RAID updates, dependency chases, steering packs, decision records. Verification and a close-out finish the work item.
A second workflow handles escalations: a blocked milestone, a failed dependency, a scope dispute. It assesses, lays out the options and who decides, waits for your approval of a response, then records what was communicated and what is being watched.
Everything the pack knows lives in a program-manager/ folder in your project: a journal with the next step, the documents, and a memory of facts and rules about your role. Close the session, switch agents, come back next week, and "continue" picks up at the exact step where the work stopped.
What makes it different:
- "Continue" is never taken as approval. A gate clears only when you say "I approve the charter" (or the plan, the change, the response).
- The agent does not accept a milestone itself when a named approver is recorded.
- Changes to scope, date, or a dependency go through a change request, not a quiet edit of the plan.
- It remembers. The pack starts with ten facts and ten rules for the role and adds your own; people-sensitive notes stay on the work item unless you say otherwise.
- Skill names are prefixed
pgm-, so the pack installs beside the other StreamTeem packs.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Business & Operations.
Works on local Markdown files only; it does not connect to Jira, Asana, or other tracking tools. Viewing the diagram page needs internet access, because the page loads Mermaid from a CDN. Python 3 is needed only to run the optional memory analysis script; without it the agent does the same counts by reading the files.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 4 days ago
- Software
- Writable project folder
- Optional
Python 3 (memory health check only)
"Continue" is never approval. Gates clear only when you approve or skip them. State lives under program-manager/ so any session can resume. Python 3 is needed only for the optional memory health check.
Permissions
Read/Write create the journal, charters, plans, diagrams, trackers, and notes under one folder in the project. Shell runs only the bundled read-only memory analysis script, and only when you ask for a memory check.