agentleFS
tatsoft-llctatsoft-llc/FrameworX-Projects/.claude/skills/new-project-bootstrap/SKILL.md

new-project-bootstrap

Skill1 starsChanged 26 days agoApache-2.0View on GitHub

Set up a new FrameworX automation project from scratch via a short interview. Use at the start of a NEW project, when the user says "set up a project for my <plant/skid/line>", "new FrameworX project", "start a solution for ...", "bootstrap this project". Creates Sites/<Name>/ from Templates/SiteProject/, interviews for site/scope, hardware/IO, and requirements; writes Specifications/Requirements.md + Specifications/IO-Register.md; creates the FrameworX solution into Solutions/ via the Designer/Console MCP; seeds Documentation/Commissioning-Record.md and the per-author ChangeLogs/; then hands off to UNS authoring and commissioning. Do NOT use for an existing project that already has a folder under Sites/. Go straight to authoring there.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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
  1. Very specific:

    Ask for the project / site name.

    line 32
  2. Distinctive:

    Ask in one batch; accept partial answers:

    line 41
  3. Very specific:

    Create the FrameworX solution via create_solution, name it after the project, target the project's Solutions/ folder (or the user's normal …

    line 69
  4. Distinctive:

    Don't over-collect. A small skid needs the IO register + a handful of requirements.

    line 82
  5. Distinctive:

    If something the project needs isn't reachable through the MCP, say so plainly.

    line 85

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

new-project-bootstrap

Turns a request into a started project folder under Sites/<Name>/, captured IO list, written requirements, a created solution, and a commissioning skeleton, in one guided pass.

Before you start: bootstrap the platform

Your first action is ALWAYS to connect to the installed FrameworX through the Designer (or Console) MCP (list_solutions / open_solution / create_solution). That returns the platform mental model (Four Pillars, UNS, object schemas, naming, syntax). Everything below assumes it is loaded. Never paraphrase platform knowledge from memory: the product just gave it to you.

Step 1: Create the project folder

  • Ask for the project / site name. Copy the Templates/SiteProject/ skeleton into Sites/<Name>/, then rename the *.template.md files to their working names (Specifications/Requirements.md, Specifications/IO-Register.md, Documentation/Commissioning-Record.md). In ChangeLogs/, rename ChangeLog-AUTHOR.template.md to ChangeLog-<your-login>.md: one file per contributor, per KnowledgeBase/Conventions/ChangeLog-Convention.md.
  • Open your ChangeLogs/ChangeLog-<login>.md with a first dated entry: "Project bootstrapped."

Step 2: Project & scope (short)

Ask in one batch; accept partial answers:

  • One-line description of what the project controls or monitors.
  • Deployment size: small-but-critical / medium / large. (Sets rigor, not a different process.)
  • Target FrameworX version (default: the installed one).
  • Who signs off commissioning (name / role).

→ Set the model tier: balanced for routine; escalate to most-capable for large or safety/availability-critical scope (see CLAUDE.md → "Model tiers").

Step 3: Hardware / IO

Either interview point-by-point or invite the user to paste their IO list / instrument index in any format and normalize it. Per point (leave unknowns blank, do not invent): instrument/tag ID, description, signal type (AI/AO/DI/DO), protocol & device/channel, address, units, range, scaling, alarm limits, and the target UNS tag path (PascalCase, area/equipment hierarchy; confirm with the user, check KnowledgeBase/Conventions/UNS-Naming.md for the house convention).

→ Write into Sites/<Name>/Specifications/IO-Register.md.

Step 4: Requirements

Capture briefly: purpose; functional behaviors (loops, sequences, interlocks); operational needs (Displays, alarms, trends, reports, roles); KPIs; acceptance criteria.

→ Write into Sites/<Name>/Specifications/Requirements.md.

Step 5: Create the solution & seed commissioning

  • Create the FrameworX solution via create_solution, name it after the project, target the project's Solutions/ folder (or the user's normal solutions location; record where in Solutions/README.md). This is the customer's REAL solution, not a sandbox.
  • Seed Documentation/Commissioning-Record.md with one row per IO point (status: pending).

Step 6: Confirm, log & route

  • Summarize what was created (IO points, requirements sections, solution name/path) and append it to your ChangeLogs/ChangeLog-<login>.md.
  • Route to next: author the UNS from the IO register → build Displays → commission point-by-point (drive each tag with set_object_value, read back value + quality + timestamp, record in the commissioning record).

Notes

  • Don't over-collect. A small skid needs the IO register + a handful of requirements; a large plant fills more. Capture what the user has, mark the rest TODO, move on.
  • A trick that will help future projects → capture it in KnowledgeBase/, not buried in one project.
  • If something the project needs isn't reachable through the MCP, say so plainly.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.