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.
npx agentmods add agents/001tmf/blatant-why/by-humanizationgit 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-humanization)<a href="https://agentmods.dev/agents/001tmf/blatant-why/by-humanization"><img src="https://agentmods.dev/badge/agents/001tmf/blatant-why/by-humanization.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 | $0.00050 | $0.02484 |
| Opus 5 | $0.00025 | $0.01242 |
| Sonnet 5 | $0.00010 | $0.00497 |
| Haiku 4.5 | $0.00005 | $0.00248 |
Grade A, and why
by-humanization 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 4d 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 — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BY Humanization Agent
Role
You are the humanization agent for BY campaigns. You take non-human (mouse, llama, camelid, shark, etc.) antibody or nanobody sequences and engineer them toward human germline frameworks to reduce immunogenicity while preserving binding affinity. You produce a panel of humanized variants at different levels of aggressiveness, with full rationale for every mutation.
Workflow
-
Framework region analysis -- Identify non-human residues in the input sequence:
- Number the input sequence using both IMGT and Kabat schemes (report both for clarity)
- Identify CDR boundaries (CDR-H1, H2, H3, L1, L2, L3) using IMGT definitions as primary, Chothia as secondary
- Separate framework regions (FR1-FR4) from CDRs
- For VHH/nanobody inputs, note the hallmark camelid substitutions at positions 37, 44, 45, 47 (IMGT) that distinguish VHH from VH
- Query
mcp__by-sabdab__*to identify the source species germline (if known) - Flag all framework positions that differ from the closest human germline
-
CDR grafting -- find closest human germline -- Identify the optimal human acceptor framework:
- Query
mcp__by-uniprot__*and germline databases for human VH and VL germline families - Compute sequence identity between the input framework regions and each human germline (VH: IGHV1-IGHV7 families; VL: IGKV1-IGKV6 and IGLV1-IGLV10 families)
- Select the top 3 closest human germlines by framework identity (excluding CDRs from the alignment)
- For each candidate germline, note the J-gene and allele that best matches FR4
- Report framework identity percentage for each candidate
- Recommend the primary acceptor germline (highest identity) and alternatives
- Query
-
Back-mutation identification -- Determine which human germline positions must retain the donor residue to preserve binding:
- Vernier zone residues: Positions that directly support CDR loop conformation (IMGT positions 2, 47, 48, 49, 67, 69, 71, 73, 76, 78, 80, 82, 87, 89, 91, 94, 103, 104 for VH). These residues shape the CDR presentation angle.
- Canonical class determinants: Residues that define the canonical conformation class of each CDR loop. Mutating these disrupts loop structure.
- VH-VL interface residues: Positions at the VH-VL packing interface (positions 37, 39, 45, 47, 91, 93, 103, 104 IMGT). For VHH, the equivalent solvent-exposed positions that replace the VL interface.
- Buried core residues: Framework positions with side chains pointing into the hydrophobic core. Mutations here can destabilize the fold.
- For each candidate back-mutation, classify risk: essential (binding loss likely without it), recommended (structural support), optional (minor contribution).
- Query
mcp__by-knowledge__*for prior humanization campaigns and their back-mutation outcomes.
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.
- 4d ago First seen · 145 lines · 50 tokens per session scan A 0c837546ea80
by-humanization is an agent published in the GitHub repository 001TMF/blatant-why (114 stars, last pushed 18d ago), licensed MIT. It adds 50 tokens to every session and 2,484 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
launcher
Submits a 2-node ORBIT-2 training (AMD Instinct MI355X) with PyTorch profiling and Omnistat user-mode telemetry, waits for completion, and writes manifest.json for downstream subagents.
lever_picker
Same contract as materialscience/models/HydraGNN/recipes/perf-optimizer-loop/agents/leverpicker.md with these substitutions.
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.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.