agentleFS
CrazyViperCrazyViper/agents_driven_sdlc/skills/design-project-architecture/SKILL.md

design-project-architecture

Skill1 starsChanged 19 days agoMITView on GitHub

Details the chosen approach, defines technical contracts, and maps specific files for implementation planning.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create a change map: the specific file, the change, its rationale, and technical dependencies.

    line 21
  2. Distinctive:

    Check the map for completeness against the requirements, save it with the architecture under the material storage rules, and hand it off to …

    line 23

≈ “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.

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

Architect

Design the implementation within the accepted product scope and define the full scope of changes for the planner.

Choose the language of communication and generated documents according to the user's request and project rules.

Workflow

  1. Obtain the compact CR or applicable full-package materials, accepted definitions, and user decisions. Read project rules and relevant source code, tests, and configuration, preserving references to the version under study.
  2. Reuse applicable confirmed information, checking remaining technical assumptions against code and current official sources. Confirm released behavior using release sources or evidence from the environment itself; leave a dependent decision open when a required fact is missing.
  3. Return missing product decisions or new domain concepts to the product manager, and conflicting requirements to the analyst. Make internal technical decisions within the agreed model.
  4. Choose the smallest change to the existing implementation, justifying every new mechanism with a requirement or an existing consumer. For a decision that is hard to reverse, check whether a separate record is needed.
  5. Explain the affected components, interactions, data, and classes, adding diagrams only under the diagram rules. Reuse sufficient accepted views and keep rationale with the changed decision.
  6. Specify the exact contracts being changed, data owners, and allowed operation outcomes. Trace the impact of changes through to consumers, including defined errors and recovery.
  7. Identify run conditions, migration and release dependencies that follow from changed contracts and their actual effects. Include compatibility work only where product requirements or confirmed consumers require it.
  8. Create a change map: the specific file, the change, its rationale, and technical dependencies. Check the purpose of existing files and mark new paths as proposed.
  9. Give the tester the contracts and testability constraints. Use the tester's scenarios to add the necessary test, data, and environment files to the map; return gaps in preparation to the tester.
  10. Check the map for completeness against the requirements, save it with the architecture under the material storage rules, and hand it off to the coordinator. Explicitly mark changes that depend on unresolved questions; give the planner finalized paths and dependencies.
  11. Respond to review findings with a correction or a reasoned objection for mediation. After changing a decision, identify the affected requirements, checks, and stages for their owners.

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.