block-structure-resolver

A procedure for finding the established content-table format for a named website block. A block is a reusable section of a web page, and the procedure also lists its known variations.

In plain words
What is it for?
It is for resolving the exact authoring structure and variants from test examples, shared pages, or, when necessary, the block's code.
Why use it?
It prevents authors from inventing a table structure or variation name that the existing code does not support.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/adobecom/da-express-milo/block-structure-resolver
Clone the repo
git clone --depth 1 https://github.com/adobecom/da-express-milo

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,409 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.02409
Opus 5 $0.00000 $0.01205
Sonnet 5 $0.00000 $0.00482
Haiku 4.5 $0.00000 $0.00241

Measured 2d ago against content hash accc7b3885d5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

block-structure-resolver scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s "https://main--da-express-milo--adobecom.aem.live<path>.plain.html"
.claude/skills/author-page-from-figma/agents/block-structure-resolver.md · 217 lines

How it starts

The opening of the file, as written. The whole thing — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Block Structure Resolver Subagent

Delegated from Phase 2 of the main SKILL.md. Given a confirmed block name, resolves the exact DA authoring table structure for that block plus its list of known variants — without inventing or generalizing a structure. This skill never defines authoring structure; it always copies one that already exists.


Inputs

  • Block name (confirmed in Phase 1)
  • Repo root: /Users/meganthomas/Projects/da-express-milo

Resolution cascade

Try each source in order. Each source yields the same two outputs: a base table structure and a variant registry. Record which source was used — it is reported to the user in the final summary and affects confidence.

The cascade is per-variant, not per-block — don't stop early just because an earlier source answered something. Nala answering the base variant does not mean it answered the specific variant Phase 4 needs. If the variant you need isn't in the source that succeeded, keep walking the remaining sources for that variant specifically before concluding it's code-inference-only. This mattered in practice: a block's base variant came from its nala fixture, but the one variant actually needed only existed in that same block's kitchen-sink page — checking kitchen sink only ever happens if nala's answer is treated as per-variant, not as "this block is resolved, move on."

Source 1 — Nala test fixtures (preferred)

Nala's block.json is a structured registry, not just markup: each entry names a variant, the CSS selector that produces it, and a live page that already renders it. This is the only source that answers "what variants exist and what does each look like" directly, so prefer it whenever it exists.

  1. Check for the fixture:

    ls nala/blocks/<block-name>/<block-name>.block.json
    

    If absent, go to Source 2.

  2. Read the file. For each entry in variants, extract:

    • name (e.g. @cards-dark)
    • selector (e.g. div.cards.dark) — the class list is the ground truth for what a variant is actually called in code. Some entries have no selector; infer the variant token from the name suffix instead (e.g. @cards-featuredfeatured).
    • path (e.g. /drafts/nala/blocks/cards-featured)

Read the full file on GitHub · 217 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 217 lines · 0 tokens per session scan A accc7b3885d5

Subscribe to this mod's changes

block-structure-resolver is an agent published in the GitHub repository adobecom/da-express-milo (6 stars, last pushed 3d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 2,409 tokens. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.