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 skills add SPYfighting/my-academic-skills --skill binder-design-campaigngit clone --depth 1 https://github.com/SPYfighting/my-academic-skillsWrote 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/skills/spyfighting/my-academic-skills/binder-design-campaign)<a href="https://agentmods.dev/skills/spyfighting/my-academic-skills/binder-design-campaign"><img src="https://agentmods.dev/badge/skills/spyfighting/my-academic-skills/binder-design-campaign/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.
<a href="https://agentmods.dev/skills/spyfighting/my-academic-skills/binder-design-campaign"><img src="https://agentmods.dev/badge/skills/spyfighting/my-academic-skills/binder-design-campaign.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00148 | $0.01976 |
| Opus 5 | $0.00074 | $0.00988 |
| Sonnet 5 | $0.00030 | $0.00395 |
| Haiku 4.5 | $0.00015 | $0.00198 |
Grade A, and why
binder-design-campaign 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 3d 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 — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gated protein design campaign
Three branches, one core. Pick the branch at G0; the gates, evidence rules and deliverable discipline below hold for all three.
| Branch | Read | For |
|---|---|---|
| binder design | workflows/binder-design.md |
miniprotein binders against a protein target |
| structure prediction | workflows/structure-prediction.md |
folds or complexes with calibrated confidence |
| protein engineering | workflows/protein-design-general.md |
stability, enzyme, interface, scaffold redesign |
Provenance and rationale: docs/ and NOTICE.md, relative to this skill.
Non-negotiables
These hold at every depth, including inside sub-agents, and bind hardest on the steps nobody is watching.
- State only what you established. Every claim in a report, update or manifest traces to an executed computation or a saved artifact you can point at. Verified means you ran the check and hold its output.
- Every external identifier is fetched. DOIs, PMIDs, PDB IDs, UniProt accessions, residue ranges and URLs in any deliverable are the literal output of a lookup executed in this session.
- Anomalies are bugs until investigated. A score of exactly zero, a gate that passes everything, a gate that fails everything, a constant metric, a perfect metric, an impossible runtime, zero variance across seeds: halt that stage, diagnose, then proceed.
- Cheapest falsifying check first. Run the seconds-scale check that could kill a plan before committing hours of GPU to it.
- Lead with the unfavourable reading. Headline the worst defensible interpretation of your own data. Disclose deviations from the stated method, and name analyses tuned after seeing results along with how many configurations you tried. Report inconclusive results as inconclusive.
- The instrument is a filter, never an oracle. In silico confidence ranks
designs within a target. It does not tell you whether a target will work,
and it relates only weakly to affinity — measured, not assumed
(
references/failure-modes.md§1). Every summary you write says so. - Verify inherited inputs. Check target structures, construct definitions and hotspot numbering against their primary source before building on them.
- Scope. This protocol is in silico and it ends at the design sheet.
Validation is in vitro, by the user or a CRO; analysing returned assay data
is a different task and does not run under these gates. Results from an
earlier round enter here as operator input at G0, never as something this
protocol goes and fetches. Work on select agents or toxins, and enhancement
of pathogen function, is out of scope: stop and say so. See
docs/responsible-use.md.
What ships with it
38 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- checklists/gate-checklists.md 14 KB
- docs/design-rationale.md 7.7 KB
- docs/responsible-use.md 2.8 KB
- docs/source-mapping.md 10 KB
- examples/design_sheet.csv 6.3 KB
- examples/ledger/demo-worker.jsonl 910 B
- examples/ledger/rejects/demo-worker.jsonl 1.2 KB
- examples/README.md 2.5 KB
- examples/sheet_schema.json 7.9 KB
- examples/state/gates/EGFR.json 160 B
- examples/state/gates/PD-L1.json 161 B
- examples/state/governor.json 87 B
- examples/state/heartbeat.json 70 B
- examples/wallclock_hours.json 28 B
- LICENSE 1.1 KB
- NOTICE.md 5.8 KB
- references/compute-backends.md 12 KB
- references/deliverables-spec.md 14 KB
- references/failure-modes.md 9.0 KB
- references/scoring-instrument.md 13 KB
- references/target-dossier-guide.md 7.4 KB
- references/thresholds.md 7.2 KB
- references/tool-catalog.md 16 KB
- references/validation-gates.md 7.4 KB
- scripts/check_design_sheet.py 14 KB runs code
- scripts/check_diversity.py 5.8 KB runs code
- scripts/ledger.py 15 KB runs code
- scripts/make_scoreboard.py 8.2 KB runs code
- scripts/preflight.py 4.1 KB runs code
- scripts/selfcheck.py 7.7 KB runs code
- templates/campaign_config.yaml 4.4 KB
- templates/report_outline.md 4.7 KB
- templates/scoring_policy.yaml 3.4 KB
- templates/sheet_schema.json 6.9 KB
- templates/target_dossier.md 3.1 KB
- workflows/binder-design.md 10 KB
- workflows/protein-design-general.md 6.3 KB
- workflows/structure-prediction.md 4.5 KB
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.
- 3d ago Changed 9a4a177d69d6
- 12d ago First seen · 160 lines · 148 tokens per session scan A db7c470a3235
binder-design-campaign is a skill published in the GitHub repository SPYfighting/my-academic-skills (14 stars, last pushed 4d ago), licensed MIT. It adds 148 tokens to every session and 1,976 once invoked, about $0.0007 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 2026-08-31.
Other skills, from other repositories
figure-design
Use when creating, updating, or reviewing a research figure (plot, chart, brain map, heatmap, schematic) — guides chart-type selection, typography, colorblind-safe palette, layout, and journal-specific export specs. Dispatches a figure-reviewer subagent after rendering. Complements claims-audit (which checks figure…
hypothesis-first
Use when running any experiment or analysis — hypothesis and analysis plan must be registered before seeing results, the scientific equivalent of test-driven development.
manuscript-writing
Use when writing a research manuscript — guides section-by-section writing with prerequisite gates, citation discipline, number traceability, and per-section subagent review. Activated after experiment results are reviewed and before claims-audit.
claims-audit
Use during or after manuscript writing to verify every quantitative claim traces to actual data, every figure is script-generated, and all results including negative ones are reported.
experiment-design
Use when a research design has been approved and needs to be broken into executable experiment steps — the research equivalent of an implementation plan.
research-brainstorming
You MUST use this before any experiment, analysis, or scientific claim — when a research question, hypothesis, study design, or scientific approach needs to be defined or refined.