plan-portfolio-automation
Skill1 starsChanged 3 months agoMITView on GitHub
Design the automation behind the portfolio cycle — status intake, validation, data flow to the reporting surface — with triggers, validation rules, idempotency, error escalation and a manual fallback. Automates collection and validation, never judgment. Produces a buildable spec, not a built flow. Dispatches the portfolio-analyst.
In this repository30 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- build-business-caseSkill
- build-roadmapSkill
- capture-feedbackSkill
- close-projectSkill
- coordinate-programSkill
- define-metricsSkill
- define-strategySkill
- design-dashboardSkill
- elicit-requirementsSkill
- groom-backlogSkill
- make-deliverableSkill
- manage-changeSkill
- manage-resourcesSkill
- plan-capacitySkill
- plan-launchSkill
- plan-projectSkill
- plan-sprintSkill
- powerbiSkill
- prioritizeSkill
- review-okrsSkill
- review-portfolio-intakeSkill
- run-ceremonySkill
- run-discoverySkill
- run-gate-reviewSkill
- setup-pmoSkill
- track-financialsSkill
- track-portfolioSkill
- track-statusSkill
- write-prdSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is 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.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
When the reporting cycle is consuming the time that analysis needs, when intake chasing is manual, when data moves between tools by …
- Very specific: line 44
Start from templates/automation-spec.md.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
/plan-portfolio-automation — Take the manual cost out of the cycle
When to use
When the reporting cycle is consuming the time that analysis needs, when intake chasing is manual, when data moves between tools by copy-paste, or when the same validation errors recur every period.
Dispatches
portfolio-analyst.
Inputs
knowledge/portfolio.md (the intake contract in practice — what's missing and how often), the latest intake
review in knowledge/portfolio/, existing dashboard specs (destination and measure definitions),
knowledge/integrations.md (available connectors and their status), knowledge/cadence.md (cycle timing).
Steps
- Map the current cycle end to end and mark where the time and the errors actually go — chasing submissions, re-keying data, fixing validation problems, rebuilding the same summary.
- Pick candidates on the rule that collection and validation automate well; judgment does not. Automate the chase, the checks, and the movement of data. Never automate interpreting a RAG or deciding to escalate.
- Design each flow: trigger → collect → validate → transform → store → publish → notify. Validate at the boundary — bad input rejected on arrival never reaches a report, and rejection messages name the field, the problem, and the fix.
- Make it idempotent. Define the dedupe key and the re-run behaviour; assume every run happens twice, and specify how a past period is safely backfilled.
- Specify error handling by failure class — retry transient, halt and alert on persistent, process-the-rest and report the rejects on partial — each escalating to a named human within a stated time. A flow that fails silently is worse than none: the report still renders, just wrongly.
- Define the human-in-the-loop points, the run log (what ran, when, record counts, failures — the source for the data-health measure), the manual fallback, and the rollout and rollback plan.
- Fill in
## Handoffwith the build route and any open question blocking implementation. Checkknowledge/integrations.mdfirst: there is no automation build capability today, so this spec is implemented by a person or another tool. State that plainly rather than implying a flow will appear.
Methods
knowledge/methods/portfolio-management.md (automation patterns, the intake contract, data health),
knowledge/methods/project-management.md (change control and risk for the rollout).
Output
Start from templates/automation-spec.md. A spec saved to knowledge/portfolio/automation-<flow>.md — a
living document, updated in place. knowledge/ remains the source of truth; no flow may become the only
record of portfolio data. Note any new tool dependency in knowledge/integrations.md and log the decision to
knowledge/decision-log.md. Follow standards/document-standards.md.
Created by Colin Beck<br>
LinkedIn: https://www.linkedin.com/in/beckcolin/<br>
GitHub: https://github.com/link7373
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.

