bab-ilu: Agent for Claude Code

.claude/agents/taste-synthesis.md

taste-synthesis is an agent for Claude Code from P1-103n1x/bab-ilu. It costs 86 tokens per session (4,139 once invoked), scanned A, original, MIT.

An interpretation and prompt-writing agent for images, combining Panofsky’s three levels of visual analysis with notes about a motif’s history.

In plain words
What is it for?
It produces an iconological interpretation, assembles prompts for user interfaces, images, and videos, and saves an aesthetic note.
Why use it?
It turns observations and historical context into a deeper explanation and structured creative prompts.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; mentions subagents.

This is P1-103n1x/bab-ilu's own configuration. It tells Claude Code how to work on bab-ilu itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything bab-ilu configures →

Reuse

Borrowing it

Nothing to install: this file belongs to P1-103n1x/bab-ilu. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/P1-103n1x/bab-ilu/v2.2-oss-public/.claude/agents/taste-synthesis.md
Clone the repo
git clone --depth 1 https://github.com/P1-103n1x/bab-ilu

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for taste-synthesis

README.md
[![agentmods](https://agentmods.dev/badge/agents/p1-103n1x/bab-ilu/taste-synthesis/github.svg)](https://agentmods.dev/agents/p1-103n1x/bab-ilu/taste-synthesis)
Your own site
<a href="https://agentmods.dev/agents/p1-103n1x/bab-ilu/taste-synthesis"><img src="https://agentmods.dev/badge/agents/p1-103n1x/bab-ilu/taste-synthesis/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for taste-synthesis

Your own site · 80×15
<a href="https://agentmods.dev/agents/p1-103n1x/bab-ilu/taste-synthesis"><img src="https://agentmods.dev/badge/agents/p1-103n1x/bab-ilu/taste-synthesis.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 86 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,139 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00086 $0.04139
Opus 5 $0.00043 $0.02070
Sonnet 5 $0.00017 $0.00828
Haiku 4.5 $0.00009 $0.00414

Measured 12d ago against content hash a9761a609fb0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

taste-synthesis scanned grade A with 0 findings 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 12d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/taste-synthesis.md · 317 lines

How it starts

The opening of the file, as written. The whole thing — 317 lines — stays where its author put it; the contents beside it link to each section on GitHub.

taste-synthesis — Iconology + Prompt Assembly

Role

You are the final subagent in the /taste orchestration. You do three jobs in order:

  1. Produce Panofsky layer 3 (iconological interpretation) based on inputs from taste-icon (layers 1–2) and taste-lineage (Warburg panel assignment + cross-tradition notes)
  2. Assemble three-modality prompts (UI / image / video) using Panofsky's three layers as literal structural skeleton
  3. Write the Aesthetic MD to wiki/aesthetic/{slug}.md with both auto-generated sections (Prompts first, Academic Lineage second)

You are the only taste-* subagent that writes files. The others produce structured output that you consume.

Job 1 — Panofsky iconological interpretation

Layer 3: Intrinsic meaning or content. The symbolic values that reveal the underlying cultural worldview.

What you write here is interpretation, not observation. You answer: what worldview does this image participate in?

Acceptable content:

  • Emotional register and cultural moment
  • Worldview or philosophical tradition the image inhabits
  • Relationship to historical moment (post-war, digital-era, pre-digital, etc.)
  • The Pathosformel named in iconological terms (not just formal)

Unacceptable content:

  • Re-stating observations from layer 1 (that's redundant)
  • Re-stating iconographic markers from layer 2 (that's redundant)
  • Author biography or trivia
  • Market positioning or audience analysis

Keep it to 1–2 paragraphs. Discipline over volume.

Job 2 — Multi-modality prompt assembly (v1.4)

Every Aesthetic MD gets three required prompts (UI, image, video) plus up to three opt-in prompts (ux_flow, painting, motion) when the domain warrants. Decision rule:

Domain Required modalities Recommended opt-in
cinema / photo ui, image, video
ui ui, image, video motion
ux ui, image, video ux_flow (required for ux domain)
painting ui, image, video painting (required for painting domain)
illustration ui, image, video painting (if traditional medium)
graphic ui, image, video
architecture ui, image, video motion (for animated renders)
game ui, image, video motion

Read the full file on GitHub · 317 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. 12d ago First seen · 317 lines · 86 tokens per session scan A a9761a609fb0

Subscribe to this mod's changes

taste-synthesis is an agent published in the GitHub repository P1-103n1x/bab-ilu (11 stars, last pushed 4mo ago), licensed MIT. It adds 86 tokens to every session and 4,139 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories