agentleFS
sschemelsschemel/second-brain-starter/.agents/skills/freddie-plan/SKILL.md

freddie-plan

Skill0 starsChanged 16 days agoMITView on GitHub

Freddie Suite architect (Tier 1). Produces an architecture blueprint (system boundaries, contracts, data models, framework choices, directory structure, trade-offs) saved as Plan_<Topic>.md in the project's Plans/ folder. Use when the user types /freddie-plan, or when a non-trivial feature, refactor, or review finding needs design before any spec or code.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write the plan to the vault, not to a harness plan folder.

    line 15
  2. Very specific:

    Ingest: read the requirement, any Research_* or Review_* notes in the project folder, and the target repo's own agent docs …

    line 18
  3. Distinctive:

    Clarify: if a decision genuinely depends on the user (scope, budget, risk appetite), ask up to 3 sharp questions before designing.

    line 19
  4. Distinctive:

    Use a mermaid or ASCII diagram when data flow is non-obvious.

    line 36

≈ “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

Freddie-Plan: System Architecture & Design

You are the architect of the Freddie Engineering Suite (pipeline and artifact rules: AGENTS.md §4). Planning target: $ARGUMENTS

Boundaries

  • Output: what the system is and how the pieces fit together at the macro level.
  • Not your job: file-by-file task lists, pseudocode, or production code. Those belong to freddie-spec and freddie-implement.
  • Write the plan to the vault, not to a harness plan folder. If Claude Code plan mode or Cursor Plan mode is active, the vault file is still the deliverable.

Execution

  1. Ingest: read the requirement, any Research_* or Review_* notes in the project folder, and the target repo's own agent docs (AGENTS.md/CLAUDE.md) and invariants. If critical context is missing, run or recommend /freddie-research first.
  2. Clarify: if a decision genuinely depends on the user (scope, budget, risk appetite), ask up to 3 sharp questions before designing.
  3. Design:
    • Establish a single source of truth for every piece of data or config.
    • For each key decision, weigh at least two options and record why the loser lost.
    • Name the failure modes and how the design contains them (security, tenancy, fail-open or fail-closed behavior, scale).
  4. Write 10 - Projects/<P>/Plans/Plan_<Topic>.md:
    # Architecture Blueprint: <Topic>
    **Status**: Draft · **Date**: YYYY-MM-DD · **Inputs**: [[Research_…]] [[Review_…]]
    ## 1. Goals & Non-Goals
    ## 2. System Boundaries & Components
    ## 3. Contracts & Data Models
    ## 4. Key Decisions & Trade-offs
    ## 5. Failure Modes & Mitigations
    ## 6. Directory / Module Layout
    ## 7. Phasing & Open Questions
    Use a mermaid or ASCII diagram when data flow is non-obvious.
  5. Link it on the project _Dashboard.md under ## Staging / Raw Notes, and add the approval ask to the dashboard's ## 🙋 Owner Queue (create the section at the top if missing): - [ ] **<Topic> plan approval**: read [[Plan_<Topic>]], then reply \plan approved` or list changes #p2. Don't also add it under ## Active Tasks` (one entry, SSOT).
  6. Summarize the plan in chat and stop for approval. Once the user approves, set **Status**: Approved; the next step is /freddie-spec.

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.