Skill complexa-target
✓ OficialAñade, edita, inspecciona o valida definiciones de objetivos para ligandos de proteínas, ligandos de moléculas pequeñas y andamiaje de motivos AME mediante archivos YAML o CLI.
- Estrellas en GitHub
- 3934
- Archivos
- 19
- Último commit
- 11 nov 2026
- Licencia
- Apache-2.0
- Herramientas permitidas
- 4
¿Qué hace la skill complexa-target?
Añade, edita, inspecciona o valida definiciones de objetivos para ligandos de proteínas, ligandos de moléculas pequeñas y andamiaje de motivos AME mediante archivos YAML o CLI.
Resumen generado a partir del SKILL.md de la skill.
Sobre la skill complexa-target
Esta SKILL permite a biólogos computacionales y desarrolladores gestionar las definiciones de objetivos para ligandos de proteínas, ligandos de moléculas pequeñas y andamiaje de motivos AME. Permite la adición, edición e inspección precisa de los registros de objetivos mediante la manipulación directa de archivos YAML o la interfaz de línea de comandos complexa target. Al interactuar con archivos de configuración específicos como targets_dict.yaml o ligand_targets_dict.yaml, los usuarios pueden definir parámetros críticos como selectores de cadenas y residuos, hotspots y cadenas SMILES de ligandos.
Un agente debe cargar esta SKILL cuando un usuario necesite registrar un nuevo objetivo de diseño o actualizar uno existente dentro de un pipeline de Proteina-Complexa. La SKILL genera bloques de configuración YAML actualizados y puede verificar la resolución de rutas utilizando el comando complexa validate target. Aunque la CLI proporciona avisos de seguridad y comillas automáticas, la edición directa de archivos es el método preferido para flujos de trabajo de agentes para evitar problemas de escape de shell. Esta SKILL funciona independientemente de una GPU en ejecución o de un entorno de ejecución instalado, centrándose en la integridad estructural de los archivos de configuración.
Lo que un agente puede hacer con ella
- 1Registrar un nuevo objetivo de ligando de proteínaAñade una nueva entrada de objetivo de proteína al registro predeterminado utilizando un nombre, un directorio de origen y un rango de residuos.
Escrito a partir del SKILL.md de la skill. Cada caso de uso se verifica con una cita del mismo.
Cómo instalar la skill complexa-target
Cada comando copia la carpeta de la skill al directorio de skills del cliente. Reinicia el cliente o inicia una nueva sesión para cargarla.Instalar la skill complexa-target en Claude Code
~/.claude/skills/complexa-target/SKILL.md
$ npx degit NVIDIA/skills/skills/bionemo-complexa-target ~/.claude/skills/complexa-targetInstalar la skill complexa-target en Claude Code (project)
.claude/skills/complexa-target/SKILL.md
$ npx degit NVIDIA/skills/skills/bionemo-complexa-target .claude/skills/complexa-targetInstalar la skill complexa-target en Codex
~/.agents/skills/complexa-target/SKILL.md
$ npx degit NVIDIA/skills/skills/bionemo-complexa-target ~/.agents/skills/complexa-targetInstalar la skill complexa-target en Gemini CLI
~/.gemini/skills/complexa-target/SKILL.md
$ npx degit NVIDIA/skills/skills/bionemo-complexa-target ~/.gemini/skills/complexa-targetInstalar la skill complexa-target en Claude.ai
npx degit NVIDIA/skills/skills/bionemo-complexa-target complexa-target && zip -r complexa-target.zip complexa-target # claude.ai → Customize → Skills → + → Create skill → Upload a skill → complexa-target.zip
Verificaciones automáticas
Ejecutado 15 nov 2026Informe completo →Aún sin verificar. Una skill necesita 80 puntos, ningún hallazgo grave y datos suficientes para comprobarla. Cómo funcionan las verificaciones · Qué corregir para obtener más puntos
- Repositorio de origen vinculado y accesiblenvidia/skills/skills/bionemo-complexa-target is public and reachablegithub.com/nvidia/skills/tree/HEAD/skills/bionemo-complexa-t ↗Aprobado
- Licencia de código abierto (OSI)Apache-2.0spdx.org/licenses/Apache-2.0.html ↗Aprobado
- Commit o lanzamiento en el último añoLast commit 2026-11-11github.com/nvidia/skills/commits ↗Aprobado
- Sin descargar y ejecutar ni robo de credenciales en los scriptsNo scripts in the skill folderAprobado
- Sin instrucciones ocultas en SKILL.mdNo override phrases, safeguard switches, credential requests or hidden charactersgithub.com/nvidia/skills/blob/HEAD/skills/bionemo-complexa-t ↗Aprobado
- Las herramientas preaprobadas están acotadasBash lets the agent run any shell command without askinggithub.com/nvidia/skills/blob/HEAD/skills/bionemo-complexa-t ↗Hallazgo
- SKILL.md sigue el formato de Agent Skillsname "complexa-target" differs from the folder "bionemo-complexa-target"github.com/nvidia/skills/blob/HEAD/skills/bionemo-complexa-t ↗Hallazgo
SKILL.md
Ver en GitHubcomplexa-target
Add or edit a design target in Proteina-Complexa. Targets live in three YAML files, one per complexa design pipeline:
configs/targets/targets_dict.yaml— protein binder (default pipeline)configs/targets/ligand_targets_dict.yaml— ligand binderconfigs/design_tasks/ame_dict_v2.yaml— AME / enzyme scaffolding
The complexa target CLI only manages the first two. AME tasks use an extended schema (the same core fields as ligand targets plus contig_atoms for hand-curated per-residue motif atom selections) and are file-edit-only.
Instructions
complexa target add is a thin wrapper around "load YAML → build dict → append a block" (see src/proteinfoundation/cli/target_manager.py:add_target_cli and append_target_to_dict). For agentic use, just edit the targets dict directly — the schema is short, the existing entries are great copy templates, and you skip 14 CLI flags / SMILES shell-escaping. Use the CLI only if you explicitly want its YAML auto-quoter or its overwrite-prompt safety.
For an offline edit, work on the supplied registry or requested output copy. When a copy is requested, create its parent directory and copy into an absent destination. If the destination already exists, inspect and preserve it: reuse it when it is the intended working copy, otherwise report the path conflict. Do not delete the output tree to restart a configuration edit. An installed Complexa runtime, GPU, or target structure is not required to write the configuration. Use the field schema below, preserve unrelated entries, and read back the saved block. Coordinate validation remains pending when the structure is absent; finish the configuration task without installing a runtime. With Read/Edit tools, read the destination copy before editing it; reading the original input path does not satisfy Edit's requirement for the output path.
Use the schema below and the selected registry; the CLI reference is needed
only for CLI use. Finish with the saved entry, its consumer pipeline, the
downstream generation.task_name, and the validation status.
| Step | Direct file edit (preferred) | CLI |
|---|---|---|
| Look at existing targets | Read configs/targets/{,ligand_}targets_dict.yaml (or configs/design_tasks/ame_dict_v2.yaml for AME) |
complexa target list (protein/ligand only) |
| Look at one target | Read the dict and grep for the key | complexa target show NAME |
| Add a new target | Append a YAML block (Step 3a) | complexa target add ... (Step 3b) |
| Verify configuration | Parse the saved YAML and compare the changed entry and untouched keys | complexa target show NAME --dict PATH |
| Check runtime PDB-path resolution | Requires the full pipeline and environment | complexa validate target CONFIG --target NAME |
Step 1: Decide the target type
Targets live in three different files, one per complexa design pipeline. Pick the file by working out which pipeline the user will run, then add the entry to that file. Each file is consumed by exactly one pipeline.
| User intent | Pipeline | Targets dict file | This skill? |
|---|---|---|---|
| Bind a protein surface (PD-L1, IFNAR2, TNF-α, …) — default | configs/search_binder_local_pipeline.yaml |
configs/targets/targets_dict.yaml |
Yes — Step 2 (protein) |
| Bind a small-molecule pocket (FAD, SAM, OQO, …) | configs/search_ligand_binder_local_pipeline.yaml |
configs/targets/ligand_targets_dict.yaml |
Yes — Step 2 (ligand) |
AME / enzyme scaffolding (motif + ligand, M####_<pdb> names) |
configs/search_ame_local_pipeline.yaml |
configs/design_tasks/ame_dict_v2.yaml |
Partial — see "AME tasks" below |
AME tasks (enzyme pipeline)
AME tasks live under motif_target_dict_cfg: with source, target_filename,
ligand, contig_atoms, binder_length, and use_bonds_from_file. The
contig_atoms string contains curated chain/residue/atom selections. Existing
entries illustrate the field structure, not the selections for a new motif.
If motif atoms, ligand selection, or structure location are missing, list the
missing inputs and write a separate, explicitly incomplete template if useful.
Leave the registry unchanged until those scientific inputs are supplied. A task
name or PDB ID alone does not determine them. In the missing-input handoff,
explain that registration will be a direct edit under motif_target_dict_cfg
once the curated selections are supplied; complexa target add cannot register
AME tasks. Completed entries are selected with ++generation.task_name=<NAME>
and configs/search_ame_local_pipeline.yaml.
Step 2: Gather required info
Use the supplied values and ask only for missing required information. Required fields differ for protein, ligand, and AME definitions.
Protein target
| Field | Question | Example | Required |
|---|---|---|---|
| name | "Target name (used as the dict key and task_name)?" |
02_PDL1, MyTarget_v1 |
yes |
| source | "Source directory under $DATA_PATH/target_data/?" |
bindcraft_targets, custom_targets |
yes (or target_path) |
| target_filename | "PDB filename (no .pdb extension)?" |
PD-L1, IFNAR2 |
yes (or target_path) |
| target_input | "Chain + residue range — see reference for grammar." | A1-115, A1-50,B1-50 |
yes |
| hotspot_residues | "Hotspot residues (interface contact residues)?" | ["A33", "A95", "A102"] |
optional, recommended |
| binder_length | "Binder length range [min, max] or single [length]?" |
[80, 150], [100] |
optional (default [60, 120]) |
| pdb_id | "Reference PDB ID (optional, metadata only)?" | "2lag" |
optional |
Hotspot selectors encode chain + residue number, not amino-acid identity.
For example, Tyr42 (Y42) on explicitly specified chain B becomes B42;
already supplied chain/residue selectors stay unchanged. Preserve the user's
residue numbering. Without coordinates, report that residue identities and
membership in the structure are unverified.
Ligand target — protein fields above (minus target_input), plus:
| Field | Question | Example | Required |
|---|---|---|---|
| ligand | "3-letter PDB ligand residue code?" | FAD, OQO, SAM |
yes (presence marks target as ligand) |
| SMILES | "SMILES string for the ligand?" | "O=C2C3=Nc1cc(c(...)..." |
recommended; uppercase YAML key |
| ligand_only | "Generate pocket around ligand only (no protein-protein interface)?" | true / false |
optional (default true) |
| use_bonds_from_file | "Use bond info from the input PDB/CIF?" | true / false |
optional (default true) |
| target_input | not required for ligand targets | — | no |
Check whether the proposed key already exists in the selected YAML mapping. Preserve an existing entry unless the user requested that entry's update.
Copy a supplied SMILES exactly. Chain-specific ligand selection belongs to
preparation of the input structure; do not invent a chain field or protein
target_input to represent it in the ligand entry.
Examples
Step 3a: Append the YAML block directly (preferred)
Open configs/targets/targets_dict.yaml (or ligand_targets_dict.yaml for ligand targets), find a similar existing entry as a style template, and append the new block under target_dict_cfg:. Two-space indent, single blank line between entries.
Protein template
02_PDL1:
source: bindcraft_targets
target_filename: PD-L1
target_input: "A1-115"
hotspot_residues: ["A37", "A39", "A49", "A98"]
binder_length: [64, 155]
pdb_id: "4z18"
Rules to match the existing file style (mirrors what complexa target add would emit):
- Quote chain/residue ranges (
"A1-115","A33") and any SMILES — flow-style strings get tripped up by:and brackets otherwise. - Use flow-style lists (
["A33", "A95"],[64, 155]) — that's what the on-disk dump produces. target_input,source,target_filenameare required for protein.target_path(absolute path) can replacesource + target_filenameif the PDB lives outside$DATA_PATH/target_data/.
Ligand template
41_7BKC_LIGAND:
source: ligand_targets
target_filename: 7BKC_ligand_centered
hotspot_residues: [null]
binder_length: [100]
pdb_id: 7BKC
ligand: 'FAD'
ligand_only: True
SMILES: "O=C2C3=Nc1cc(c(cc1N(C3=NC(=O)N2)CC(O)C(O)C(O)COP(=O)(O)OP(=O)(O)OCC6OC(n5cnc4c(ncnc45)N)C(O)C6O)C)C"
use_bonds_from_file: True
The presence of the ligand: key flips the target into ligand mode — there is no separate is_ligand flag. target_input is not required for ligands.
After saving, skip to Step 4 to verify.
Step 3b: CLI alternative (complexa target add)
Use the CLI when you want its automatic chain/residue quoting, the overwrite-confirm prompt, or to wire target creation into a non-Python script. Prefer non-interactive mode for agentic use (-i / --editor opens an editor and blocks).
Confirmed flags (from src/proteinfoundation/cli/target_cli.py)
Read target_schema.md for the flag
table and worked CLI examples when this path is needed. Select the intended
dictionary with --dict; --source and --target-filename should be explicit
when they differ from the CLI defaults. AME registration remains a direct edit.
Step 4: Verify
First parse the saved YAML with an available YAML parser and check the selected mapping, field types, requested values, and preservation of other entries. Read back the resulting block. This verifies configuration editing only.
With a prepared Complexa installation, pipeline configs, environment, and target structure, additionally check runtime path resolution from that checkout:
# 1. Select the dictionary that was actually edited.
complexa target show 02_PDL1 --dict configs/targets/targets_dict.yaml
# 2. Resolve the PDB path and validate it exists on disk.
# This traverses Hydra defaults to find the target dict.
complexa validate target configs/search_binder_local_pipeline.yaml --target 02_PDL1
complexa validate target (from src/proteinfoundation/cli/validate.py::validate_target) checks:
DATA_PATHis set andtarget_data/exists.- The target name resolves in
target_dict_cfg. - Either
target_pathexists, or$DATA_PATH/target_data/<source>/<target_filename>.pdbexists. target_input,hotspot_residues,binder_lengthare reported back so a human can sanity-check them.
For ligand targets, use configs/targets/ligand_targets_dict.yaml for show
and configs/search_ligand_binder_local_pipeline.yaml for validation; --target
is the target key, not a config path. This validator does not check residue
identities, ligand chemistry, or AME contig_atoms; do not describe a path check
as coordinate or scientific validation.
Step 5: Emit artifact
Save the edited entry and validation status at the user's requested output
location (otherwise ./target_<name>/). For an offline edit, the saved YAML
block is sufficient; no CLI-generated artifact is required. On a prepared
installation, a target-show record can also be saved:
mkdir -p target_02_PDL1
complexa target show 02_PDL1 > target_02_PDL1/target_show.txt
If a standalone definition is requested, save the actual entry under its
registry key in target_definition.yaml. The handoff should include the saved
path, target key, consumer pipeline from Step 1, and
++generation.task_name=<NAME>, together with what validation actually ran.
For an incomplete AME request, report the separate template, missing inputs,
and the direct registry edit needed after those inputs arrive.
Hardware requirements
None for target definition — this is a YAML edit, not a training/inference step. Disk impact is a few KB appended to the targets dict (plus an automatic .yaml.bak backup written by save_targets_dict).
For the downstream design / evaluate runs that consume the target, defer to complexa-design and references/hardware.md.
Troubleshooting
| Symptom | Cause | Fix |
|---|---|---|
Target PDB file: File not found from validate target |
<source>/<target_filename>.pdb does not exist under $DATA_PATH/target_data/ |
Confirm the PDB stem and source dir. Use --target-path /full/path.pdb if the file lives outside target_data/. |
| "Target 'X' already exists! Overwrite? (y/N)" | Name collision in dict | Either pick a new name, or pass -f / --force to overwrite. |
| Hotspot residue not in PDB | Wrong chain or residue number | Open the PDB, re-check chain letters (case-sensitive) and residue indices. Hotspots use the format <CHAIN><RESNUM> — see reference. |
| Chain not found | target_input references a chain that does not exist in the PDB |
Inspect the PDB with grep "^ATOM" target.pdb | awk '{print $5}' | sort -u. |
| Ligand code missing from PDB | The 3-letter ligand code does not appear as a HETATM residue name in the file |
Open the PDB and check HETATM lines; you may need a _ligand_centered variant of the PDB. |
| SMILES parse failure downstream | Bad SMILES string | Validate with rdkit.Chem.MolFromSmiles(smiles) before adding. Quote SMILES in shell to escape brackets and parens. |
| Target name with leading digit breaks Hydra interpolation | YAML treats 02_PDL1 as a string fine; some override syntaxes need quoting |
Use ++generation.task_name=02_PDL1; quote if the shell strips characters. |
target_input appears ignored for a ligand target |
By design — ligand targets do not use target_input; pocket is defined by the ligand |
Leave it unset for ligand targets. |
Reference
references/target_schema.md— every field, chain-spec grammar, AME task-name grammar, three worked examples.configs/targets/targets_dict.yaml— live protein entries (copy a known-good one as a template).configs/targets/ligand_targets_dict.yaml— live ligand entries.configs/design_tasks/ame_dict_v2.yaml— AME task definitions (file-edit only, not exposed viacomplexa targetCLI).src/proteinfoundation/cli/target_cli.py— argparse source of truth.src/proteinfoundation/cli/target_manager.py—add_target_cli,list_targets,show_target, schema inTARGET_FIELDS.src/proteinfoundation/cli/validate.py—validate_targetimplementation.
Archivos de esta skill
19Preguntas sobre la skill complexa-target
¿Cómo instalo la skill complexa-target en Claude Code?
Ejecuta npx degit NVIDIA/skills/skills/bionemo-complexa-target ~/.claude/skills/complexa-target. Claude Code carga las skills desde ~/.claude/skills (todos los proyectos) o .claude/skills (un proyecto) y usa una skill cuando su descripción coincide con la tarea.
¿Qué clientes pueden usar la skill complexa-target?
Cualquier cliente que lea el formato Agent Skills: Claude Code, Codex, Gemini CLI y claude.ai (subir como ZIP). Los scripts de la skill se ejecutan solo donde el cliente permite la ejecución de código.
¿Cómo añado un nuevo objetivo de proteína?
El método preferido es añadir un bloque YAML directamente al archivo 'configs/targets/targets_dict.yaml' utilizando los campos obligatorios como name, source, target_filename y target_input.