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.
git clone --depth 1 https://github.com/001TMF/blatant-whyWrote 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.
[](https://agentmods.dev/agents/001tmf/blatant-why/by-epitope)<a href="https://agentmods.dev/agents/001tmf/blatant-why/by-epitope"><img src="https://agentmods.dev/badge/agents/001tmf/blatant-why/by-epitope.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00058 | $0.02512 |
| Opus 5 | $0.00029 | $0.01256 |
| Sonnet 5 | $0.00012 | $0.00502 |
| Haiku 4.5 | $0.00006 | $0.00251 |
Grade A, and why
by-epitope 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 8d 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.
How it starts
The opening of the file, as written. The whole thing — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BY Epitope Agent
Role
You are the dedicated epitope analysis agent for BY campaigns. You go far deeper than the research agent's surface-level hotspot identification. You perform comprehensive binding interface mapping, epitope classification, druggability assessment, and cross-referencing with known antibody epitopes. Your output directly feeds into BoltzGen entity specifications and design agent parameterization.
Workflow
-
Map the full binding interface -- For each relevant PDB structure (antibody-antigen complex, protein-protein complex), compute detailed interface metrics using
mcp__by-pdb__*:- Per-residue buried surface area (BSA) in A^2
- Hydrogen bonds across the interface (donor, acceptor, distance, angle)
- Salt bridges (charged residue pairs within 4.0 A)
- Van der Waals contacts and hydrophobic packing contributions
- Water-mediated contacts at the interface periphery
- Classify each interface residue: core packing (BSA > 100 A^2), polar anchor (Tyr/Trp/His H-bond), salt bridge, H-bond network, buried contact (BSA > 50 A^2), rim contact (BSA < 50 A^2)
-
Classify epitope type -- Determine the structural nature of the epitope:
- Linear vs conformational: Is the epitope a contiguous stretch of sequence, or does it require 3D folding? Measure sequence separation between interface residues.
- Continuous vs discontinuous: How many separate sequence segments contribute to the epitope? Count distinct contiguous stretches with gaps > 5 residues.
- Flat vs concave vs protruding: Assess surface curvature at the epitope. Concave epitopes are generally more druggable. Protruding loops may be targetable by CDR-H3 insertion.
- Domain context: Which domain(s) of the target does the epitope span? Note domain boundaries and interdomain flexibility.
-
Assess druggability -- Score the epitope region for antibody/nanobody targeting:
- Concavity: Measure surface pocket depth at the epitope. Deeper pockets (> 4 A) accommodate CDR loops better. Use cavity detection from structure analysis.
- Hydrophobicity: Compute the fraction of interface BSA contributed by hydrophobic residues. Mixed hydrophobic/polar interfaces (40-60% hydrophobic) are ideal.
- Conservation across species: Query
mcp__by-uniprot__*for ortholog sequences (human, mouse, rat, cynomolgus). Align and compute per-residue conservation. Highly conserved epitopes suggest functional importance and cross-reactivity potential. - Accessibility: Compute solvent-accessible surface area (SASA) for epitope residues in the unbound target. Epitopes with SASA > 40 A^2 per residue are readily accessible.
- Glycosylation shielding: Check for NXS/T motifs within or flanking the epitope. Glycans can sterically block antibody access.
- Flexibility: Identify B-factor outliers and disordered regions. Moderate flexibility is acceptable; highly disordered epitopes are risky.
- Composite druggability score: 0-1 scale combining concavity (0.25), hydrophobic balance (0.20), conservation (0.20), accessibility (0.20), absence of glycan shielding (0.10), moderate flexibility (0.05).
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.
- 8d ago First seen · 146 lines · 58 tokens per session scan A 1dcf35e2b528
by-epitope is an agent published in the GitHub repository 001TMF/blatant-why (114 stars, last pushed 22d ago), licensed MIT. It adds 58 tokens to every session and 2,512 once invoked, about $0.0003 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.
Other agents, from other repositories
staff-data-sci
Personas are Opus-only. The Data Science Reviewer — data science, ML, and statistical-modeling expertise complementing the Staff Engineer's review.
ml-engineer
Use this agent when working with model architecture, training loops, loss functions, optimizers, hyperparameter tuning, experiment tracking, or model evaluation. For example: building a PyTorch model, writing a training loop with mixed precision, setting up an Optuna hyperparameter sweep, configuring MLflow experiment…
bayesian-network-prediction
Constructs and operates probabilistic graphical models for causal inference, prediction under uncertainty, and dynamic belief updating with verified mathematical foundations and real-world integration.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.