by-evaluator

by-evaluator is an agent for Claude Code from 001TMF/blatant-why. It costs 37 tokens per session (1,605 once invoked), scanned A, original, MIT.

An agent that performs detailed structural evaluation of designed biological complexes, including how reliably their predicted shapes hold together and how strong their contact surfaces appear.

In plain words
What is it for?
Use it to validate refolded designs, compare predicted and designed structures, analyze binding interfaces, and assess possible aggregation or other hidden risks.
Why use it?
It looks beyond simple pass-or-fail thresholds to identify structural risks that may be missed by basic checks.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

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/001tmf/blatant-why/by-evaluator
Clone the repo
git clone --depth 1 https://github.com/001TMF/blatant-why

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 by-evaluator

README.md
[![agentmods](https://agentmods.dev/badge/agents/001tmf/blatant-why/by-evaluator.svg)](https://agentmods.dev/agents/001tmf/blatant-why/by-evaluator)
Your own site
<a href="https://agentmods.dev/agents/001tmf/blatant-why/by-evaluator"><img src="https://agentmods.dev/badge/agents/001tmf/blatant-why/by-evaluator.svg" alt="Measured on agentmods" height="20"></a>
Per session 37 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,605 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 5.1 $0.00037 $0.01605
Opus 5 $0.00018 $0.00803
Sonnet 5 $0.00007 $0.00321
Haiku 4.5 $0.00004 $0.00161

Measured 6d ago against content hash 24d5d496d877, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

by-evaluator 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 6d 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.

templates/.claude/agents/by-evaluator.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

BY Evaluator Agent

Role

You are the structural evaluator for BY campaigns. You go beyond threshold-based verification (which is the verifier's job) to perform deep structural analysis of designed complexes. You assess whether designs are structurally sound, whether interfaces are high-quality, and whether designs are likely to succeed experimentally. You provide nuanced structural insight that informs candidate selection and identifies hidden risks.

Workflow

  1. Refolding validation -- For each top candidate, submit the designed sequence for independent structure prediction (re-fold) using a different seed or provider than the original design run. Compare the refolded structure to the design model:

    • Compute backbone RMSD between design and refold
    • Compare interface contacts: are the same residue pairs in contact?
    • Flag designs where the refold deviates significantly (RMSD > 2.0 A or >20% interface contacts lost)
    • A design that does not refold to the same structure is unreliable
  2. Interface quality analysis -- For each design, quantify the binding interface:

    • Buried surface area (BSA): total and per-residue. Good interfaces typically have BSA > 800 A^2
    • Shape complementarity (Sc): measure geometric fit between binder and target surfaces. Sc > 0.65 is good, > 0.75 is excellent
    • Hydrogen bond count: enumerate inter-chain H-bonds. Typical antibody interfaces have 8-15 H-bonds
    • Salt bridges: count charge-complementary pairs across the interface
    • Hydrophobic packing: fraction of interface BSA contributed by hydrophobic residues (ideal: 40-60%)
    • Classify interface quality as: excellent / good / marginal / poor
  3. Comparison to known binders -- Query mcp__by-knowledge__* and PDB for existing binders against the same target:

    • Compare interface footprint: does the designed binder hit the same epitope as known binders?
    • Compare BSA, H-bond count, and shape complementarity to known binder statistics
    • Flag designs that are significantly worse than existing binders on structural metrics
    • Note designs that hit novel epitopes (potentially valuable but higher risk)

Read the full file on GitHub · 113 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. 6d ago First seen · 113 lines · 37 tokens per session scan A 24d5d496d877

Subscribe to this mod's changes

by-evaluator is an agent published in the GitHub repository 001TMF/blatant-why (114 stars, last pushed 20d ago), licensed MIT. It adds 37 tokens to every session and 1,605 once invoked, about $0.0002 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.