Skip to content
Back to skills

Plan To Blueprint

ASecurity

Transform a technical plan into an executable project skill blueprint: entry skill, command routing, workflow contracts, manifests, and runbook links.

  • 72 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added June 5, 2026
toolsgo

Security analysis

A100/100

Scanned September 30, 2026

npx -y skills add devton/agentic-workflow-blueprint --skill plan-to-blueprint --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Plan To Blueprint?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Plan To Blueprint
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/devton-plan-to-blueprint/badge)](https://www.skillsdirectory.com/skills/devton-plan-to-blueprint)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
## agentic-workflows-blueprint.workflow.plan-to-blueprint

### Goal

Transform a technical plan into an executable project skill blueprint: entry skill, command routing, workflow contracts, manifests, and runbook links.

### Scope

- Applies to: converting structured implementation plans into `skills/<projectSlug>/` scaffold outputs.
- Does not cover: implementing product features described in the plan.

### Triggers

- "Turn this plan into a skill blueprint"
- "Convert plan to executable workflows"
- "Scaffold skills from this technical plan"
- After a Plan-mode session when the user wants persistent, rerunnable agent contracts

### Inputs

- `projectSlug`: short identifier for the target repo
- `skillName` (optional): slash name for router invocation; defaults to `projectSlug`
- `baseBranch`: default integration branch
- `techStack`: short stack description
- `existingRootDoc`: root instruction file path (`AGENTS.md`, `CLAUDE.md`, etc.)
- `technicalPlan`: structured plan (milestones, tasks, acceptance criteria, constraints)
- `workflowsWanted`: workflow ids inferred from the plan (must include at least one)
- `constraints`: project hard rules extracted from the plan
- `operationalContext` (optional): environment tiers, maintenance window, rollback strategy, compliance boundaries

### Invariants

- `SKILL.md` remains the source of truth; `template.json` mirrors routing, never replaces contracts.
- Router-only command model: `/<skillName> <cmd>` maps to `workflows/<cmd>/SKILL.md`.
- Every workflow in `workflowsWanted` gets a full contract (Goal through References).
- Do not invent scope not present in `technicalPlan`; mark gaps explicitly in handoff notes.
- Workflow ids and folder names must match exactly (`kebab-case`).
- For infra/network/IaC/OS plans, include explicit operational verification and rollback criteria in generated contracts.

### Procedure

1. Parse `technicalPlan` and extract: objectives, milestones, deliverables, constraints, and candidate workflow boundaries.
2. Propose `workflowsWanted` (confirm or refine with user if ambiguous).
   - Include `network-engineering`, `infra-operations`, `iac`, and `os-platform` when the technical plan includes those domains.
3. Scaffold `skills/<projectSlug>/SKILL.md` with orchestrator + **Command routing** (router-only).
4. Generate `skills/<projectSlug>/template.json` with `commands[]` for each workflow in `workflowsWanted`.
5. Create `reference/routing-matrix.md` and `reference/role-contracts.md` from plan evidence.
6. For each workflow id, create `workflows/<workflowName>/SKILL.md` with executable contract sections grounded in the plan.
7. Optionally add `workflows/<workflowName>/template.json` when a tool needs per-workflow manifests.
8. Wire `existingRootDoc` with "Start here" and skill/runbook links (no duplication of workflow bodies).
9. Add or update runbook `docs/runbooks/plan-to-blueprint.md` when operator guidance is needed.
10. Run consistency verification: links, ids, manifest commands, routing-matrix alignment.

### Outputs

- `skills/<projectSlug>/SKILL.md` (entry orchestrator with command routing).
- `skills/<projectSlug>/template.json` (declarative command manifest).
- `skills/<projectSlug>/reference/routing-matrix.md` and `role-contracts.md`.
- `skills/<projectSlug>/workflows/<workflowName>/SKILL.md` for each item in `workflowsWanted`.
- Updated `existingRootDoc` with canonical links.
- Handoff note listing mapped plan sections -> workflows and any unresolved gaps.

### Review gate

- [ ] Every command in `template.json` resolves to an existing `workflows/<cmd>/SKILL.md`.
- [ ] Entry skill documents router-only invocation (`/<skillName> <cmd>`).
- [ ] Each workflow contract is complete and traceable to plan evidence.
- [ ] Root doc links only to canonical paths; no duplicated workflow content.
- [ ] All referenced paths exist and ids follow `<projectSlug>.workflow.<name>`.

### References

- `../../SKILL.md`
- `../document/SKILL.md`
- `../review/SKILL.md`

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.