<< All versions
Skill v1.0.1
currentAutomated scan100/100eightmm/codex-science/cx-boltz-structure-prediction
+2 new
──Details
PublishedNovember 5, 2026 at 08:14 PM
Content Hashsha256:5ac35ca48f010c8e...
Git SHA79a408e0987d
Bump Typepatch
──Files
Files (1 file, 2.2 KB)
SKILL.md2.2 KBactive
SKILL.md · 38 lines · 2.2 KB
name: cx-boltz-structure-prediction description: "Predict 3D biomolecular structures and binding affinity with the open-source Boltz model — proteins, complexes, nucleic acids, and protein-ligand systems from sequence. Use when the user wants a predicted structure or affinity from sequence (not an experimental structure or a database lookup). Downloads and installs Boltz into an isolated uv environment and runs it." metadata: version: "1.0.1"
<!-- Generated by scripts/generate_wrappers.py; do not edit. -->
Cx Boltz Structure Prediction
Use this Codex adapter for the pinned upstream scientific skill.
Catalog status
- Status:
active - Source:
cx - Audit reasons: none
- Upstream license:
MIT - Inventory:
../../inventory.json - Upstream instructions:
../../../authored-skills/boltz-structure-prediction/SKILL.md
Procedure
- Resolve the plugin root as three directories above this skill directory.
- Run
"<plugin-root>/scripts/python_runtime.sh" "<plugin-root>/scripts/search_skills.py" --name cx-boltz-structure-predictionand verify the record's status, reasons, and source path and the source's pinned commit (empty for first-partycxskills) from its output instead of readinginventory.json. - Read the upstream instructions only after confirming this record is still active in the current inventory.
- Treat upstream frontmatter as source metadata, not as Codex configuration. Follow the instruction body only where it is compatible with available Codex tools and the current user-approved plan.
- Treat tools named for another agent runtime as capability requests. Map them to an available Codex tool only when the behavior is equivalent; otherwise report the unavailable capability.
- Ask before credentials, package installation, new network hosts, remote compute, writes, paid services, destructive actions, or imported executable code.
- Use
$science-provenancefor produced scientific outputs and$science-reviewbefore presenting scientific claims.
Boundaries
- This wrapper does not grant credentials, network access, package installation, or script execution.
- Do not change an inactive record to active from inside a research task.
- Preserve failures, uncertainty, source identifiers, and licensing constraints.