@delta-agentcloud/cli 0.0.0-stage → 1.1.20

package/skills/delta-protein/references/plip_analyze.md

@@ -0,0 +1,65 @@
+# PLIP Protein-Ligand Interaction Analysis (`huada_protein_tools` / `plip_analyze`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+Analyze interactions between a protein and a ligand with PLIP.
+- Input: `structure_file` (PDB/CIF file URL)
+- Optional:
+- `ligand_id`: ligand identifier, e.g. "LIG" or "A:LIG:301"; leave empty and PLIP auto-detects the largest ligand
+- `interaction_types`: list of interaction types to return, e.g. ["hydrogen_bonds", "hydrophobic_interactions"]
+- Output: interaction results `resultjson`, `summary`, `download_url`
+- Use case: identify protein-ligand interactions such as hydrogen bonds, hydrophobic contacts, salt bridges, pi-pi stacking, halogen bonds, metal coordination
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint plip_analyze --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `structure_file`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `structure_file` | PDB/CIF | - |
+| `ligand_id` | Ligand identifier, e.g. LIG or A:LIG:301; empty = auto-detect | - |
+| `interaction_types` | List of interaction types to return; default returns all | - |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: derived from ToolSpec/agent templates; pending live calibration
+
+- Projection content: Protein-ligand interactions: hydrogen bonds, hydrophobic contacts, salt bridges, pi-pi stacking etc. + summary + download_url
+- Presentation form: Interaction-category text/table + links
+- Downstream reuse: input is often the rank-1 pdb_url returned by dock
+## Output Template
+
+- Title: **PLIP protein-ligand interaction analysis result**
+- Lead with the conclusion: detected interaction categories (hydrogen bond/hydrophobic/salt bridge/pi-pi/halogen/metal coordination) and key residues
+- Add result files and visualizations to the "Result files & downloads" section; note the source structure when the input came from dock
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).

package/skills/delta-protein/references/pocket_prediction.md

@@ -0,0 +1,63 @@
+# Pocket Prediction (`huada_protein_tools` / `pocket_prediction`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+Predict protein pockets with p2rank (local).
+- Required: `structure_file` (.pdb/.cif)
+- Optional: `model_config_name` (default | alphafold), default "default"; AlphaFold/NMR/cryo-EM structures are best served by "alphafold"
+- Output: `resultjson` contains `n_pockets` and a `pockets` list (rank/name/score/probability/center/residue_ids/n_residues); `download_url` is the cropped pocket PDB download link ({structure_name}_pocket_cut.pdb, pocket residues + ligand HETATM only), suitable as RFdiffusion3 design input
+- Use case: identify functionally important residues and ligand binding sites
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint pocket_prediction --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `structure_file`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `structure_file` | PDB/CIF | - |
+| `model_config_name` | p2rank model configuration: default or alphafold | default |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: verified by a real `delta-cli` invocation
+
+- Projection content: n_pockets and pockets list (rank/name/score/probability/center/residue_ids/n_residues) + cropped pocket PDB download_url
+- Presentation form: Pocket list + link; n_pockets=0 is a valid result (p2rank found no pocket), not a failure
+- Downstream reuse: cropped pocket PDB (download_url) -> structure_file for rfd3_design
+## Output Template
+
+- Title: **Binding pocket prediction result**
+- Lead with the conclusion: number of pockets, per-pocket score/probability, location of the top ones (center and residues)
+- Add the cropped pocket PDB to the "Result files & downloads" section and note it can feed rfd3_design
+- Report n_pockets=0 truthfully (p2rank found no pocket), not a failure
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).

package/skills/delta-protein/references/protein_design.md

@@ -0,0 +1,65 @@
+# Protein Design (RFdiffusion3+MPNN) (`huada_protein_tools` / `protein_design`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+De novo protein design with RFdiffusion3 and MPNN.
+- Required: `rfd3_json_url` (RFdiffusion3 config file)
+- Optional: `input_file` (list of PDB files), `mpnn_type` (protein_mpnn/ligand_mpnn/soluble_mpnn), `rfd3_n` (1), `mpnn_n` (1)
+- Output: designed sequences with scores `download_url`, `zip_download_url`
+- Use case: de novo design of new proteins; use when the user uploads both a PDB and an RFdiffusion3 JSON, or explicitly mentions rfd3/RFdiffusion3
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint protein_design --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `rfd3_json_url`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `rfd3_json_url` | RFdiffusion3 config file | - |
+| `input_file` | Input PDB file URL list | [] |
+| `mpnn_type` | MPNN type, default protein_mpnn | protein_mpnn |
+| `rfd3_n` | RFdiffusion3 sample count, default 1 | 1 |
+| `mpnn_n` | MPNN sample count, default 1 | 1 |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: derived from ToolSpec/agent templates; pending live calibration
+
+- Projection content: Designed sequences with scores, download_url (design results), zip_download_url
+- Presentation form: Download link list; structure files given as `[name](full URL)` per format
+- Downstream reuse: terminal artifact
+## Output Template
+
+- Title: **Protein design result (RFdiffusion3 + MPNN)**
+- Lead with the conclusion: number of designed sequences / scores
+- Add design-result and scaffold-file links to the "Result files & downloads" section: `[Design result file](download_url)`, `[ZIP](zip_download_url)`
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).

package/skills/delta-protein/references/protein_evolution.md

@@ -0,0 +1,77 @@
+# Protein n-site Mutation Activity Prediction (`huada_protein_tools` / `protein_evolution`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+Predict the activity of specified mutation sites with a protein-evolution method.
+- Required: `mutant_url` (Excel/CSV file with a mutant column and at least one numeric column)
+- Optional: `protein_sequence`/`sequence_file`/`structure_file`, `chain` (A), `mutant_n` (0), `topk` (-1), `model_name` (esm2/esmc), `threshold`, `regression_type` (randomforest/gp)
+- Multi-objective optimization: `target_columns` (specify multiple numeric column names, e.g. ["yield","ee"]; auto-detected by default), `weights` (per-target weights; equal weights by default)
+- Output: best mutations with activity scores `download_url`
+- Use case: mutation activity data is available and the user explicitly specifies "2-site/3-site/n-site/multi-site mutation"
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint protein_evolution --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `mutant_url`
+
+Server-side default parameters:
+
+- `topk` = `30`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `mutant_url` | CSV/XLSX with a mutant column and at least one numeric column | - |
+| `protein_sequence` |  | - |
+| `sequence_file` | FASTA/FA | - |
+| `structure_file` | PDB/CIF | - |
+| `chain` | Protein chain identifier, default A | A |
+| `mutant_n` | Specified number of n-site mutations, default 0 | 0 |
+| `topk` | Number of results to return, default -1 (all) | -1 |
+| `model_name` | Feature model name (esm2/esmc) | - |
+| `threshold` | Threshold | - |
+| `regression_type` | Regression model (randomforest/gp) | - |
+| `target_columns` | Specify multiple target numeric column names, e.g. ['yield','ee']; auto-detects all numeric columns by default | - |
+| `weights` | Weights per target column, e.g. [0.5,0.5]; equal weights by default | - |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: derived from ToolSpec/agent templates; pending live calibration
+
+- Projection content: Activity score topk for specified n-site/multi-site mutations (download_url)
+- Presentation form: Text/table + download link
+- Downstream reuse: -
+## Output Template
+
+- Title: **Protein n-site mutation activity prediction result**
+- Lead with the conclusion: n-site/multi-site mutation activity prediction results (topk)
+- Add details to the "Result files & downloads" section
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).

package/skills/delta-protein/references/protein_sol_predict_mutation.md

@@ -0,0 +1,66 @@
+# Mutant Solubility Prediction (`huada_protein_tools` / `protein_sol_predict_mutation`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+Predict the solubility of mutant proteins.
+- Input: `protein_sequence`/`sequence_file`/`structure_file` + `mutant_list`/`mutant_list_file`
+- Optional: `chain` (A), `min_sol` (-1.0)
+- Output: solubility prediction results per mutant `download_url`
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint protein_sol_predict_mutation --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `protein_sequence` / `sequence_file` / `structure_file`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `protein_sequence` |  | - |
+| `sequence_file` | FASTA/FA | - |
+| `structure_file` | PDB/CIF | - |
+| `chain` | Protein chain identifier, default A | A |
+| `mutant_list` | Mutation list | [] |
+| `mutant_list_file` | CSV/XLSX mutation list file | - |
+| `min_sol` | Minimum solubility threshold, default -1.0 | -1.0 |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: derived from ToolSpec/agent templates; pending live calibration
+
+- Projection content: Solubility score per mutant and change trend; flag mutations likely to affect solubility
+- Presentation form: Text/table
+- Downstream reuse: -
+## Output Template
+
+- Title: **Mutant solubility prediction result**
+- Lead with the conclusion: solubility score/change trend, flagging mutations likely to affect solubility
+- Add details to the "Result files & downloads" section
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).

package/skills/delta-protein/references/protein_sol_predict_protein.md

@@ -0,0 +1,61 @@
+# Protein Solubility Prediction (`huada_protein_tools` / `protein_sol_predict_protein`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+Predict protein solubility and isoelectric point (pI).
+- Input: `protein_list` / `protein_list_file` (choose one)
+- Important: `protein_list_file` must be a `.fasta/.fa/.csv/.xlsx` URL; never pass structure files
+- Output: solubility and pI results `download_url`
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint protein_sol_predict_protein --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `protein_list` / `protein_list_file`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `protein_list` | List of single or multiple one-letter amino-acid sequences | [] |
+| `protein_list_file` | FASTA/FA/CSV/XLSX sequence file | - |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: derived from ToolSpec/agent templates; pending live calibration
+
+- Projection content: Protein solubility and isoelectric point pI (download_url)
+- Presentation form: Text
+- Downstream reuse: -
+## Output Template
+
+- Title: **Protein solubility prediction result**
+- Lead with the conclusion: predicted solubility trend and pI
+- Add details to the "Result files & downloads" section
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).

package/skills/delta-protein/references/protenix.md

@@ -0,0 +1,66 @@
+# Protein Structure Prediction (Protenix) (`huada_protein_tools` / `protenix`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+Predict the three-dimensional structure of a protein with the Protenix model.
+- Input: `protein_sequence` / `sequence_file` (choose one)
+- Optional: `use_msa` (false)
+- Output: predicted structure file (PDB) `download_url`
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint protenix --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `protein_sequence` / `sequence_file`
+
+Server-side default parameters:
+
+- `use_msa` = `false`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `protein_sequence` |  | - |
+| `sequence_file` | FASTA/FA | - |
+| `use_msa` | Whether to use multiple sequence alignment | false |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: verified by a real `delta-cli` invocation
+
+- Projection content: Predicted structure download_url (.cif/.pdb) with confidence: plddt, ptm, gpde, num_recycles, has_clash
+- Presentation form: Structure file link + confidence text
+- Downstream reuse: structure can serve as downstream structure_file, or extract the sequence first via structure_to_sequence and use as protein_sequence
+## Output Template
+
+- Title: **Protenix structure prediction result**
+- Lead with the conclusion: prediction done + confidence (plddt/ptm/gpde), num_recycles, clash status
+- Add the structure file link to the "Result files & downloads" section: `[Protenix structure prediction result file](download_url)`
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).

package/skills/delta-protein/references/protrek_search.md

@@ -0,0 +1,90 @@
+# ProTrek API Search (`huada_protein_tools` / `protrek_search`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+Open-ended protein retrieval (text/sequence/structure) across multiple databases
+with the ProTrek protein search API.
+- Typical scenarios:
+- Search related proteins by functional/reaction/family descriptions in protein
+  databases (`input_type=text`).
+- Search similar proteins by sequence (`input_type=sequence`).
+- Search similar structures by structure (`input_type=structure`).
+- Not for: exact metadata queries of a known UniProt accession / UniParc ID / gene
+  name / protein name (use `uniprot_query` for those).
+- Optional parameters: `input_type` (text/sequence/structure), `query_type`
+  (sequence/structure), `topk` (10), `db` (Swiss-Prot), `protein_sequence`,
+  `sequence_file`, `structure_file`, `chain` (A), `text`
+- `query_type` defaults to `sequence`;
+- when the user mentions "structure search / structure / 3D structure", choose `structure`;
+- `input_type` is inferred from whether a sequence/structure file is provided; when
+  neither is provided, `input_type` is `text`;
+- the `text` parameter is specifically for functional/reaction/family
+  description-style text searches;
+- Supported databases: with `query_type=sequence`, Swiss-Prot/UniRef50/Uncharacterized/OMG_prot50/PDB/GOPC/NCBI/MGnify/OMG are supported; with `query_type=structure`, only PDB
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint protrek_search --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `protein_sequence` / `sequence_file` / `structure_file` / `text`
+
+Server-side default parameters:
+
+- `input_type` = `sequence`
+- `query_type` = `sequence`
+- `topk` = `10`
+- `db` = `Swiss-Prot`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `input_type` | Input type: text/sequence/structure; inferred from whether a sequence/structure file is provided, text when neither | sequence |
+| `query_type` | Query type: sequence/structure; default sequence; choose structure when "structure search" is mentioned | sequence |
+| `topk` | Number of results to return, default 10 | 10 |
+| `db` | Database name or list, default Swiss-Prot; with query_type=sequence supports ['Swiss-Prot', 'UniRef50', 'Uncharacterized', 'OMG_prot50', 'PDB', 'GOPC', 'NCBI', 'MGnify', 'OMG']; with query_type=structure only PDB | Swiss-Prot |
+| `protein_sequence` |  | - |
+| `sequence_file` | FASTA/FA | - |
+| `structure_file` | PDB/CIF | - |
+| `chain` | Protein chain identifier, default A | A |
+| `text` | Core search description when input_type=text, e.g. function/reaction/family; translate to English | - |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: derived from ToolSpec/agent templates; pending live calibration
+
+- Projection content: Hit protein entries, similarity and functional information
+- Presentation form: Hit list text/table
+- Downstream reuse: for known targets with metadata use uniprot_query; this endpoint is for open-ended retrieval
+## Output Template
+
+- Title: **ProTrek database search result**
+- Lead with the conclusion: number of hit protein entries; list similarity per hit (score vs query) and functional information
+- For structure/sequence search, state the retrieval mode (query_type/input_type)
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).

package/skills/delta-protein/references/pubchem_search.md

@@ -0,0 +1,68 @@
+# PubChem Compound Search (`huada_protein_tools` / `pubchem_search`)
+
+This endpoint is a real endpoint of the `huada_protein_tools` tool. It can only be
+called via `delta-cli science invoke`; the business response described below lives
+under the CLI top-level `data` — read fields as documented in this file, do not
+recursively search. This file only documents the endpoint schema; it does not
+authorize direct HTTP. Endpoint names are authoritative as returned by
+`delta-cli science endpoints list huada_protein_tools`; use them verbatim — never
+guess, truncate, or alias.
+
+## Overview
+
+Call PubChem (NCBI) to search basic chemical molecule information.
+- Supports querying compounds by name, SMILES, InChI, InChIKey, PubChem CID.
+- `input_type` options: name (default) | smiles | inchi | inchikey | cid.
+- `query` passes the corresponding identifier per input_type: name like "aspirin",
+  SMILES like "CC(=O)OC1=CC=CC=C1C(=O)O", CID like "2244".
+- Optional: `max_results` (default 10, max 100)
+- Output: compound results `resultjson`, containing
+  cid/name/molecular_formula/molecular_weight/canonical_smiles/inchi/inchikey/
+  iupac_name/synonyms/pubchem_url
+
+## Invocation
+
+```text
+delta-cli science invoke --tool huada_protein_tools --endpoint pubchem_search --data JSON
+```
+
+Required input groups (at least one per group; all groups must be satisfied):
+
+- `query`
+
+Request fields:
+
+| Field | Description | Default |
+|---|---|---|
+| `query` | Query string: name, SMILES, InChI, InChIKey, or CID | - |
+| `input_type` | Query type: name \| smiles \| inchi \| inchikey \| cid, default name | name |
+| `max_results` | Maximum number of compounds to return, default 10, max 100 | 10 |
+
+## Response
+
+- CLI top-level envelope: a call succeeds only when the exit code is 0 AND `ok` is `true`; the service response lives under `data`.
+- Server-side contract: `status` (success/error/partial), `message`, `tool_type`, `resultjson`, `download_url`; some endpoints also return `total_count`. `resultjson` may be a single object or an array — read only the keys documented in this file, do not recursively search.
+- Binary/file artifacts (such as the PDB/XLSX/FASTA/ZIP files that `download_url` points to) are downloaded and verified with
+  `delta-cli science file download --tool huada_protein_tools --endpoint ... --output <path>`;
+  do not treat file contents as ordinary JSON results.
+
+## Output Projection
+
+> Verification status: verified by a real `delta-cli` invocation
+
+- Projection content: Compound entries: name, synonyms, molecular_formula, molecular_weight, canonical_smiles, inchi, inchikey, cid, pubchem_url
+- Presentation form: Text (formula/weight/name/SMILES) + entry URL
+- Downstream reuse: canonical_smiles -> smiles / primary_smiles for catpred_predict / catapro_predict
+## Output Template
+
+- Title: **PubChem compound search**
+- Lead with the conclusion: compound name, molecular formula, molecular weight
+- Also give: canonical_smiles / InChIKey, cid, `[PubChem entry](pubchem_url)` (URL in a safe form)
+- Avoid listing every JSON key
+
+## Boundaries & Failure Modes
+
+- File-type parameters (`structure_file`/`sequence_file`/`msa_file` etc.) take server-accessible URL/path strings, not local upload paths; a `download_url` or `pdb_url` returned by an upstream endpoint can be backfilled directly as downstream input.
+- Make only the minimal calls needed to complete the task; `--data` carries only the fields declared by this endpoint. Do not invent enum values or branches beyond the required combinations.
+- Do not pass off local tool (RDKit/pymatgen/LAMMPS/self-attached MCP) results as this endpoint's results; do not connect directly to Huada or gateway URLs.
+- On failure, report a short summary of the CLI's raw error; do not blindly retry after remote computation timeouts (the outcome may be unknown because the task may already have happened).