agentleFS
vrajpal-jhalavrajpal-jhala/langgraph-harness/.claude/skills/sync-architecture/SKILL.md

sync-architecture

Skill32 starsChanged 28 days agoMITView on GitHub

Check whether the work just done on this branch changes anything docs/architecture.md describes (harness structure, tool surface, data flow, or a Built/Partial/Pipeline status), and update the diagram and status prose in place if so. Use when the user says "sync architecture doc", "update architecture diagram", "/sync-architecture" — or when invoked by the update-changelog skill after it scopes a change touching harness/memory/ops structure.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. ayoubben18/ab-method★ 193same name
    2. vlv-code/meta-project★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository8 other files it gives its agents.

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:

    If invoked from the update-changelog skill, reuse the diff range it already computed — don't recompute it.

    line 18
  2. Distinctive:

    Only two kinds of change are "worthy" of a doc update:

    line 29
  3. Very specific:

    Mermaid diagram: add/remove/rename nodes and edges as needed.

    line 46
  4. Distinctive:

    Phasing note: update only if the change affects sequencing (something planned for Phase 2 shipped in Phase 1, a stated prerequisite got …

    line 51
  5. Distinctive:

    Don't restructure sections, rename harnesses, or change the diagram's overall layout beyond what the change actually requires.

    line 54

≈ “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

Root docs/architecture.md is the source of truth for langgraph-harness's system architecture: four harnesses (MR Review, Work Item Resolve, Task Resolve, Chat), the Memory layer, and LLM Ops, rendered as a mermaid diagram with 🟩 Built / 🟨 Partial / ⬜ Pipeline status classes, plus matching "Status breakdown" prose and a "Phasing note".

1. Scope the change

If invoked from the update-changelog skill, reuse the diff range it already computed — don't recompute it.

Otherwise, diff since the last commit that touched the doc:

git log --oneline -- docs/architecture.md | head -1
git log --oneline <that-commit>..HEAD
git diff <that-commit>..HEAD

Only two kinds of change are "worthy" of a doc update:

  • Structural — a component (node) added or removed, a harness's tool surface or data flow changed, a queue/loop/guardrail's actual behavior changed.
  • Status transition — something the doc marks ⬜ Pipeline or 🟨 Partial actually shipped (now 🟩 Built), partially shipped (⬜ → 🟨), or something built regressed.

Skip anything else — internal refactors that don't change the shape described, dependency bumps, test-only changes, prompt wording tweaks that don't change tool surface. If nothing in the diff is worthy, say so explicitly and stop.

2. Update in place

Match the doc's existing density and tone — it names exact gaps ("a hard numeric cap on comment count, which doesn't exist yet") rather than vague hand-waving. Don't soften or generalize what you write.

  • Mermaid diagram: add/remove/rename nodes and edges as needed; move a node between the built/partial/pipeline classDef groups on a status transition.
  • Status breakdown: edit the matching bullet under the harness's ### 🟩 Built / ### 🟨 Partial / ### ⬜ Pipeline section — move the bullet to its new section on a status transition rather than just rewording it in place.
  • Phasing note: update only if the change affects sequencing (something planned for Phase 2 shipped in Phase 1, a stated prerequisite got resolved or introduced).

Don't restructure sections, rename harnesses, or change the diagram's overall layout beyond what the change actually requires.

3. Commit

git status --short -- docs/architecture.md

Stage and commit just the doc, separately from the code change and from any changelog bump:

docs: sync architecture diagram

Count this subject line against this repo's 50/72-char limit (see CLAUDE.md). Don't fold in unrelated uncommitted changes.

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.