@cyanheads/protein-mcp-server 0.8.0 → 0.8.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/AGENTS.md +44 -20
- package/CLAUDE.md +44 -20
- package/README.md +76 -88
- package/changelog/0.8.x/0.8.1.md +29 -0
- package/changelog/0.8.x/0.8.2.md +24 -0
- package/changelog/template.md +9 -26
- package/dist/index.js +3 -1
- package/dist/index.js.map +1 -1
- package/dist/mcp-server/tools/definitions/analyze-collection.tool.d.ts.map +1 -1
- package/dist/mcp-server/tools/definitions/analyze-collection.tool.js +4 -16
- package/dist/mcp-server/tools/definitions/analyze-collection.tool.js.map +1 -1
- package/dist/mcp-server/tools/definitions/compare-structures.tool.d.ts.map +1 -1
- package/dist/mcp-server/tools/definitions/compare-structures.tool.js +2 -15
- package/dist/mcp-server/tools/definitions/compare-structures.tool.js.map +1 -1
- package/dist/mcp-server/tools/definitions/find-similar.tool.d.ts +5 -0
- package/dist/mcp-server/tools/definitions/find-similar.tool.d.ts.map +1 -1
- package/dist/mcp-server/tools/definitions/find-similar.tool.js +6 -0
- package/dist/mcp-server/tools/definitions/find-similar.tool.js.map +1 -1
- package/dist/mcp-server/tools/definitions/get-annotations.tool.d.ts.map +1 -1
- package/dist/mcp-server/tools/definitions/get-annotations.tool.js +2 -11
- package/dist/mcp-server/tools/definitions/get-annotations.tool.js.map +1 -1
- package/dist/mcp-server/tools/definitions/search-structures.tool.d.ts.map +1 -1
- package/dist/mcp-server/tools/definitions/search-structures.tool.js +1 -9
- package/dist/mcp-server/tools/definitions/search-structures.tool.js.map +1 -1
- package/dist/mcp-server/tools/definitions/track-ligands.tool.d.ts.map +1 -1
- package/dist/mcp-server/tools/definitions/track-ligands.tool.js +1 -8
- package/dist/mcp-server/tools/definitions/track-ligands.tool.js.map +1 -1
- package/package.json +10 -9
- package/server.json +3 -3
package/README.md
CHANGED
|
@@ -7,7 +7,7 @@
|
|
|
7
7
|
|
|
8
8
|
<div align="center">
|
|
9
9
|
|
|
10
|
-
[](./CHANGELOG.md) [](./LICENSE) [](https://github.com/users/cyanheads/packages/container/package/protein-mcp-server) [](https://modelcontextprotocol.io/) [](https://www.npmjs.com/package/@cyanheads/protein-mcp-server) [](https://www.typescriptlang.org/) [](https://bun.sh/)
|
|
11
11
|
|
|
12
12
|
</div>
|
|
13
13
|
|
|
@@ -27,9 +27,11 @@
|
|
|
27
27
|
|
|
28
28
|
---
|
|
29
29
|
|
|
30
|
-
##
|
|
30
|
+
## Overview
|
|
31
31
|
|
|
32
|
-
|
|
32
|
+
Experimental (PDB) and predicted (AlphaFold) protein structures, federated behind one surface. Search, fetch, align, compare, and annotate structures and their ligands across RCSB, AlphaFold DB, 3D-Beacons, UniProt, InterPro, and Foldseek — all keyless. Runs as a stdio process, a local Streamable HTTP server, or the public hosted endpoint above.
|
|
33
|
+
|
|
34
|
+
### Tools
|
|
33
35
|
|
|
34
36
|
| Tool | Description |
|
|
35
37
|
|:---|:---|
|
|
@@ -41,122 +43,107 @@ Seven tools spanning the structure-research arc — discover, fetch, find homolo
|
|
|
41
43
|
| `protein_analyze_collection` | Profile the PDB into distributions and trends with server-side facets — counts, histograms, timelines, and cross-tabs. |
|
|
42
44
|
| `protein_get_annotations` | Fetch UniProt features and natural variants plus InterPro domain/family memberships with GO terms. |
|
|
43
45
|
|
|
44
|
-
###
|
|
46
|
+
### Resources
|
|
47
|
+
|
|
48
|
+
| Resource | Description |
|
|
49
|
+
|:---|:---|
|
|
50
|
+
| `pdb://{entry_id}` | Experimental structure summary for a PDB entry — title, method, resolution, organism, bound ligands, and per-entity chain IDs in both the author (`authAsymIds`) and mmCIF label (`labelAsymIds`) namespaces. |
|
|
51
|
+
| `af://{uniprot}` | Predicted-structure summary for a UniProt accession from AlphaFold DB — mean pLDDT, confidence-band fractions, model URLs, and version. |
|
|
52
|
+
|
|
53
|
+
All resource data is also reachable via tools — `pdb://{entry_id}` mirrors `protein_get_structure` for `source: experimental`, and `af://{uniprot}` mirrors it for `source: predicted`. Many MCP clients are tool-only and don't surface resources; the summaries remain reachable through the tools.
|
|
54
|
+
|
|
55
|
+
## Capability reference
|
|
45
56
|
|
|
46
|
-
|
|
57
|
+
### `protein_search_structures` <sub>tool</sub>
|
|
47
58
|
|
|
48
59
|
- Free-text, protein-sequence (triggers an mmseqs2 similarity search), and organism / method / resolution filters
|
|
49
|
-
- `content_type` scopes the search to `experimental`, `predicted`, or `all` —
|
|
50
|
-
- Every hit names its `source`; experimental sequence hits expose a chainable PDB entry `id` plus the matched polymer `entityId`, with title, method, resolution, and organism enrichment
|
|
60
|
+
- `content_type` scopes the search to `experimental`, `predicted`, or `all` (default) — `all` is a genuine union, so computed models appear alongside PDB entries
|
|
61
|
+
- Every hit names its `source`; experimental sequence hits expose a chainable PDB entry `id` plus the matched polymer `entityId`, with title, method, resolution, and organism enrichment; computed models retain their complete model ID and parsed UniProt accession
|
|
51
62
|
- `start` and `limit` page through ranked results; `nextStart` is returned while another page remains
|
|
52
|
-
- Optional `facets` return a method / organism / release-year breakdown alongside the hits
|
|
63
|
+
- Optional `facets` return a method / organism / release-year breakdown alongside the hits — each dimension may be listed once and reports how many matches carry no value for it; a capped dimension is named in `notice`, with `protein_analyze_collection` (larger `bucket_limit`) as the route to the long tail
|
|
53
64
|
- Chain hit IDs straight into `protein_get_structure`
|
|
54
65
|
|
|
55
66
|
---
|
|
56
67
|
|
|
57
|
-
### `protein_get_structure`
|
|
58
|
-
|
|
59
|
-
Fetch structures with metadata and coordinate-file URLs, resolving across providers by `source`.
|
|
68
|
+
### `protein_get_structure` <sub>tool</sub>
|
|
60
69
|
|
|
61
|
-
- `source: experimental`
|
|
62
|
-
- `
|
|
63
|
-
-
|
|
64
|
-
-
|
|
65
|
-
-
|
|
66
|
-
- Records served by the RCSB entry endpoint also carry `polymerEntities` (with both `authAsymIds` and `labelAsymIds`), `ligands`, `molecularWeight`, and `releaseDate`
|
|
67
|
-
- `include_coords` inlines coordinate content, subject to the response budget: a batch over budget returns a per-structure size outline you can re-call with `sections: [ids]`, and a single file over budget is withheld with a pointer to its `coordinateUrls` (a `sections` re-call would return the same bytes). A `sections` re-call is not re-gated — `structuredContent` carries the named payload whole at any size, while the token-bounded text surface withholds anything over the budget and points at `coordinateUrls` instead of truncating it
|
|
68
|
-
- Every response carries an `attribution` block naming the upstream data licenses and citations (see [Upstream data licensing](#upstream-data-licensing))
|
|
70
|
+
- `source: experimental` batches PDB entry IDs (also resolving computed-model IDs like `AF_*`/`MA_*` from search, tagged `source: predicted` with their provider); `source: predicted` takes UniProt accessions for AlphaFold models with pLDDT/PAE; `source: best_available` takes UniProt accessions and returns the top federated model (highest-resolution experimental if one exists, else the best prediction)
|
|
71
|
+
- Per-ID partial success — unresolved IDs land in `failed[]`; `requested`/`processed` disclose IDs dropped beyond the batch cap, and every advisory (cap, failure, overflow) joins into one `notice`
|
|
72
|
+
- Records served by the RCSB entry endpoint also carry `polymerEntities` (both `authAsymIds` and `labelAsymIds`), `ligands`, `molecularWeight`, and `releaseDate`
|
|
73
|
+
- `include_coords` inlines coordinate content, subject to a response budget — an over-budget batch returns a per-structure size outline (re-call with `sections: [ids]`), and a single oversized file is withheld with a pointer to its `coordinateUrls`
|
|
74
|
+
- Every response carries an `attribution` block naming upstream data licenses and citations
|
|
69
75
|
|
|
70
76
|
---
|
|
71
77
|
|
|
72
|
-
### `protein_find_similar`
|
|
78
|
+
### `protein_find_similar` <sub>tool</sub>
|
|
73
79
|
|
|
74
|
-
|
|
75
|
-
|
|
76
|
-
- `by: sequence` runs a synchronous RCSB mmseqs2 search; `by: structure` runs an asynchronous Foldseek search against experimental and predicted databases
|
|
77
|
-
- Query from a raw one-letter sequence, a PDB ID, or a UniProt accession
|
|
78
|
-
- Both modes accept `start` with `limit` and report `totalCount`, echoing `start` and returning `nextStart` while another page remains
|
|
80
|
+
- `by: sequence` runs a synchronous RCSB mmseqs2 search; `by: structure` runs an asynchronous Foldseek search against experimental and predicted databases — query from a raw sequence, a PDB ID, or a UniProt accession
|
|
81
|
+
- Both modes accept `start`/`limit` and report `totalCount`, echoing `start` and returning `nextStart` while another page remains
|
|
79
82
|
- Foldseek targets default to `pdb100` + `afdb50`; override via `databases` (e.g. `afdb-swissprot`, `BFVD`)
|
|
80
|
-
-
|
|
81
|
-
- Each mode reads only its own controls
|
|
83
|
+
- An async job that exceeds the poll budget returns `status: computing` with a `ticketId` — re-call with `ticket_id` to resume; a completed structure search returns the same ticket so a new `start` pages the finished job
|
|
84
|
+
- Each mode reads only its own controls (`sequence`, `max_evalue`, `min_identity` under `by: sequence`; `ticket_id`, `databases` under `by: structure`) — a field the selected mode can't consume is rejected, not ignored
|
|
82
85
|
- Each hit names the engine and source database it came from
|
|
83
86
|
|
|
84
87
|
---
|
|
85
88
|
|
|
86
|
-
### `protein_track_ligands`
|
|
87
|
-
|
|
88
|
-
Ligand discovery and binding-site analysis across the PDB.
|
|
89
|
+
### `protein_track_ligands` <sub>tool</sub>
|
|
89
90
|
|
|
90
|
-
- `mode: find_ligand` resolves a name or formula to chemical component IDs with formula, weight, SMILES, and InChIKey
|
|
91
|
-
- A formula-shaped `query` matches on exact composition
|
|
92
|
-
- `mode: structures_with_ligand` returns PDB entries containing a ligand by exact component ID
|
|
93
|
-
- `mode: structures_with_ligand` accepts `start` with `limit` and returns `nextStart` while another page remains
|
|
91
|
+
- `mode: find_ligand` resolves a name or formula to chemical component IDs with formula, weight, SMILES, and InChIKey — ranked by deposition frequency, most-common match first
|
|
92
|
+
- A formula-shaped `query` matches on exact composition, spaced (`C29 H31 N7 O`) or unspaced; anything else (a component ID included) matches on name and synonyms
|
|
93
|
+
- `mode: structures_with_ligand` returns PDB entries containing a ligand by exact component ID, with `start`/`limit` paging and `nextStart` while another page remains
|
|
94
94
|
- `mode: binding_site` returns the protein residues lining a ligand's pocket in a structure, with contact distances
|
|
95
|
-
- Binding sites are experimental-only — computed from deposited coordinates
|
|
96
|
-
|
|
97
|
-
Paged RCSB results preserve the upstream order within each response. Resolution ties and changes in the live corpus mean traversal is best-effort across calls, not a stable export snapshot.
|
|
95
|
+
- Binding sites are experimental-only — computed from deposited coordinates; predicted models carry no bound ligands
|
|
98
96
|
|
|
99
97
|
---
|
|
100
98
|
|
|
101
|
-
### `protein_compare_structures`
|
|
102
|
-
|
|
103
|
-
Structural alignment of multiple structures (up to the configured `PROTEIN_MAX_COMPARE_STRUCTURES` cap) via the RCSB Structural Comparison service.
|
|
99
|
+
### `protein_compare_structures` <sub>tool</sub>
|
|
104
100
|
|
|
105
|
-
-
|
|
106
|
-
- `reference: first` aligns every structure to the first; `reference: all_pairs` computes the full pairwise matrix
|
|
107
|
-
-
|
|
108
|
-
-
|
|
109
|
-
-
|
|
110
|
-
- Re-call with a matching `{ a, b, uuid }` entry in `resume[]` (copied from a prior response's `pairs[]`) to poll a computing pair's job instead of resubmitting
|
|
111
|
-
- Returns TM-score, RMSD, and aligned-residue count per pair, plus `modeledResidues` and `coverage` — each a `[a, b]` tuple, with coverage a 0–100 percentage of that structure's own modeled-residue count
|
|
101
|
+
- Aligns 2 to the configured cap (default 10, max 25) structures per call, via `tm-align`, `fatcat-rigid`, or `fatcat-flexible`; optional per-structure `chain` restricts the alignment to a single mmCIF label chain
|
|
102
|
+
- `reference: first` aligns every structure to the first; `reference: all_pairs` computes the full pairwise matrix; a structure repeated in `structures[]` is compared once
|
|
103
|
+
- Each pair is an independent async job with per-pair partial success — a pair still computing when the poll budget elapses returns `status: computing` with a job `uuid`; a failed pair degrades only its own row
|
|
104
|
+
- Re-call with a matching `{ a, b, uuid }` entry in `resume[]` to poll a computing pair instead of resubmitting
|
|
105
|
+
- Returns TM-score, RMSD, and aligned-residue count per pair, plus each structure's `modeledResidues` and 0–100 `coverage`
|
|
112
106
|
|
|
113
107
|
---
|
|
114
108
|
|
|
115
|
-
### `protein_analyze_collection`
|
|
116
|
-
|
|
117
|
-
Profile the PDB into distributions and trends over an optional scoping query — backed by RCSB's server-side facet engine (one call, compact buckets, no row pull).
|
|
109
|
+
### `protein_analyze_collection` <sub>tool</sub>
|
|
118
110
|
|
|
119
111
|
- Group by `method`, `organism`, `polymer_type`, `resolution`, `release_year`, or `molecular_weight`
|
|
120
112
|
- One `group_by` dimension for a breakdown, or two distinct dimensions for a cross-tab (the first nests the second); a repeated dimension is rejected
|
|
121
|
-
- `interval` sets
|
|
113
|
+
- `interval` sets a histogram bin width (a number, for `resolution` or `molecular_weight`) or date-histogram period (`year`, the only one RCSB accepts) — applies to whichever requested dimension can consume that type; rejected when neither can
|
|
122
114
|
- Scope with a free-text `query`, `organism`, `method`, or `max_resolution`; `content_type` selects the structure universe
|
|
123
|
-
- `bucket_limit` caps buckets per dimension level, not per response — a cross-tab applies it separately to the parent
|
|
124
|
-
- Every dimension reports `missingValueCount` — matches
|
|
115
|
+
- `bucket_limit` caps buckets per dimension level, not per response — a cross-tab applies it separately to the parent and each nested child, up to `bucket_limit × (1 + bucket_limit)` buckets; `notice` names every capped position and `bucketsReturned` gives the realized total
|
|
116
|
+
- Every dimension reports `missingValueCount` — matches carrying no value for that attribute (e.g. a `resolution` breakdown excludes NMR entries; computed models have neither `method` nor `resolution`)
|
|
125
117
|
|
|
126
118
|
---
|
|
127
119
|
|
|
128
|
-
### `protein_get_annotations`
|
|
120
|
+
### `protein_get_annotations` <sub>tool</sub>
|
|
129
121
|
|
|
130
|
-
|
|
131
|
-
|
|
132
|
-
-
|
|
133
|
-
-
|
|
134
|
-
- Provide a UniProt accession directly, or a PDB ID — resolved to a UniProt accession via the structure's sequence cross-reference
|
|
135
|
-
- A multi-chain PDB entry can map to several accessions; the default is the deterministic lowest-author-chain pick, with the alternatives listed under `ambiguity`. Pass `chain` (an author chain ID, e.g. `A`) to select a specific one
|
|
136
|
-
- `include` scopes which annotation classes are fetched: `features`, `domains`, `variants`, or `all`
|
|
122
|
+
- UniProt features (domains, binding sites, PTMs) and natural variants, plus InterPro domain/family memberships (Pfam, PROSITE, …) with associated GO terms
|
|
123
|
+
- Provide a UniProt accession directly, or a PDB ID — resolved via the structure's sequence cross-reference
|
|
124
|
+
- A multi-chain PDB entry can map to several accessions; the default is the deterministic lowest-author-chain pick, with alternatives listed under `ambiguity` — pass `chain` (an author chain ID) to select a specific one
|
|
125
|
+
- `include` scopes which classes are fetched (`features`, `domains`, `variants`, `all`); `limit` caps each class independently (1–200, default 50), with a truncated class disclosed in `notice`
|
|
137
126
|
- Every response carries an `attribution` block naming the upstream data licenses and citations (see [Upstream data licensing](#upstream-data-licensing))
|
|
138
127
|
|
|
139
|
-
|
|
128
|
+
---
|
|
140
129
|
|
|
141
|
-
|
|
142
|
-
|:---|:---|:---|
|
|
143
|
-
| Resource | `pdb://{entry_id}` | Experimental structure summary for a PDB entry — title, method, resolution, organism, bound ligands, and per-entity chain IDs in both the author (`authAsymIds`) and mmCIF label (`labelAsymIds`) namespaces. |
|
|
144
|
-
| Resource | `af://{uniprot}` | Predicted-structure summary for a UniProt accession from AlphaFold DB — mean pLDDT, confidence-band fractions, model URLs, and version. |
|
|
130
|
+
### `pdb://{entry_id}` <sub>resource</sub>
|
|
145
131
|
|
|
146
|
-
|
|
132
|
+
- Experimental structure summary as `application/json` — title, method, resolution, organism, bound ligands, and per-entity chain IDs in both the author (`authAsymIds`) and mmCIF label (`labelAsymIds`) namespaces
|
|
133
|
+
- Mirrors `protein_get_structure` for `source: experimental`; `entry_id` is a PDB entry ID (e.g. `4HHB`)
|
|
147
134
|
|
|
148
|
-
|
|
135
|
+
---
|
|
149
136
|
|
|
150
|
-
|
|
137
|
+
### `af://{uniprot}` <sub>resource</sub>
|
|
151
138
|
|
|
152
|
-
-
|
|
153
|
-
-
|
|
154
|
-
- Pluggable auth: `none`, `jwt`, `oauth`
|
|
155
|
-
- Swappable storage backends: `in-memory`, `filesystem`, `Supabase`, `Cloudflare KV/R2/D1`
|
|
156
|
-
- Structured logging with optional OpenTelemetry tracing
|
|
157
|
-
- STDIO and Streamable HTTP transports
|
|
139
|
+
- Predicted-structure summary as `application/json` — mean pLDDT, confidence-band fractions, model URLs (`cif`/`pdb`/`bcif`), and AlphaFold model version
|
|
140
|
+
- `uniprot` accepts a UniProt accession or an AlphaFold DB entry ID (e.g. `AF-P69905-F1`); mirrors `protein_get_structure` for `source: predicted`
|
|
158
141
|
|
|
159
|
-
|
|
142
|
+
## Features
|
|
143
|
+
|
|
144
|
+
Built on [`@cyanheads/mcp-ts-core`](https://github.com/cyanheads/mcp-ts-core): stdio and Streamable HTTP transports, pluggable auth (`none` / `jwt` / `oauth`), swappable storage (`in-memory`, `filesystem`, `Supabase`, `Cloudflare KV/R2/D1`), structured logging with optional OpenTelemetry tracing.
|
|
145
|
+
|
|
146
|
+
PDB / AlphaFold-specific:
|
|
160
147
|
|
|
161
148
|
- One federated surface over experimental (PDB) and predicted (AlphaFold / 3D-Beacons) structures — search, fetch, and compare treat both universes the same
|
|
162
149
|
- Keyless across every upstream — RCSB, AlphaFold DB, 3D-Beacons, UniProt, InterPro, and Foldseek, no API keys to provision
|
|
@@ -186,7 +173,7 @@ A public instance is available at `https://protein.caseyjhand.com/mcp` — no in
|
|
|
186
173
|
}
|
|
187
174
|
```
|
|
188
175
|
|
|
189
|
-
### Self-
|
|
176
|
+
### Self-Hosted / Local
|
|
190
177
|
|
|
191
178
|
Add the following to your MCP client configuration file. No API key is required — every upstream provider is keyless.
|
|
192
179
|
|
|
@@ -247,7 +234,7 @@ MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
|
|
|
247
234
|
|
|
248
235
|
### Prerequisites
|
|
249
236
|
|
|
250
|
-
- [Bun v1.
|
|
237
|
+
- [Bun v1.4.0](https://bun.sh/) or higher (or Node.js v24+).
|
|
251
238
|
- No accounts or API keys — RCSB, AlphaFold DB, 3D-Beacons, UniProt, InterPro, and Foldseek are all public and keyless.
|
|
252
239
|
|
|
253
240
|
### Installation
|
|
@@ -286,6 +273,7 @@ All upstream providers are keyless, so the server runs out of the box with no co
|
|
|
286
273
|
| `FOLDSEEK_BASE_URL` | Base URL for the Foldseek structural-similarity search service. | `https://search.foldseek.com` |
|
|
287
274
|
| `MCP_TRANSPORT_TYPE` | Transport: `stdio` or `http`. | `stdio` |
|
|
288
275
|
| `MCP_HTTP_PORT` | Port for the HTTP server. | `3010` |
|
|
276
|
+
| `MCP_SESSION_MODE` | HTTP session mode: `stateless`, `stateful`, or `auto`. The server declares `stateless` in code; set this to override it. | `stateless` |
|
|
289
277
|
| `MCP_AUTH_MODE` | Auth mode: `none`, `jwt`, or `oauth`. | `none` |
|
|
290
278
|
| `MCP_LOG_LEVEL` | Log level (RFC 5424). | `info` |
|
|
291
279
|
| `OTEL_ENABLED` | Enable [OpenTelemetry instrumentation](https://github.com/cyanheads/mcp-ts-core/tree/main/docs/telemetry). | `false` |
|
|
@@ -333,7 +321,7 @@ The Dockerfile defaults to HTTP transport, stateless session mode, and logs to `
|
|
|
333
321
|
| `src/config` | Server-specific environment variable parsing and validation with Zod. |
|
|
334
322
|
| `src/mcp-server/tools` | Tool definitions (`*.tool.ts`). |
|
|
335
323
|
| `src/mcp-server/resources` | Resource definitions (`*.resource.ts`). |
|
|
336
|
-
| `src/services` | Provider service layer — RCSB, AlphaFold, 3D-Beacons, UniProt
|
|
324
|
+
| `src/services` | Provider service layer — RCSB (search, data, facets), AlphaFold, 3D-Beacons (best-available), UniProt (incl. InterPro/GO), Structural Comparison alignment, Foldseek, and shared HTTP/identifier/concurrency helpers. |
|
|
337
325
|
| `tests/` | Unit and integration tests mirroring `src/`. |
|
|
338
326
|
|
|
339
327
|
## Development guide
|
|
@@ -345,15 +333,6 @@ See [`CLAUDE.md`/`AGENTS.md`](./CLAUDE.md) for development guidelines and archit
|
|
|
345
333
|
- Register new tools and resources via the barrels in `src/mcp-server/*/definitions/index.ts`
|
|
346
334
|
- Wrap external API calls: validate raw → normalize to domain type → return output schema; never fabricate missing fields
|
|
347
335
|
|
|
348
|
-
## Contributing
|
|
349
|
-
|
|
350
|
-
Issues and pull requests are welcome. Run checks and tests before submitting:
|
|
351
|
-
|
|
352
|
-
```sh
|
|
353
|
-
bun run devcheck
|
|
354
|
-
bun run test
|
|
355
|
-
```
|
|
356
|
-
|
|
357
336
|
## Upstream data licensing
|
|
358
337
|
|
|
359
338
|
Structure and annotation data comes from public upstream databases, each under its own license. `protein_get_structure` and `protein_get_annotations` carry an `attribution` block on every response — the license, citation, and homepage for each source that contributed to that specific response — so the attribution obligation travels with the data to downstream consumers rather than living only here. CC BY / CC BY-SA sources require attribution on redistribution; CC0 sources are citation-only (attribution encouraged, not required).
|
|
@@ -371,6 +350,15 @@ Structure and annotation data comes from public upstream databases, each under i
|
|
|
371
350
|
|
|
372
351
|
`best_available` federates predicted models through [3D-Beacons](https://3d-beacons.org/), so the `attribution` block credits the actual contributing provider (AlphaFold DB, SWISS-MODEL, BFVD, …); a provider without a curated license entry carries a `See provider terms` fallback pointing back to 3D-Beacons rather than a fabricated license. InterPro's own domain/family classifications are CC0; the GO terms carried alongside them are separately CC BY 4.0, so each is credited independently only when it actually contributes. Full citations for each source travel in the `attribution` block of the relevant tool responses. This covers upstream *data* licensing — the server's own code is licensed separately (see [License](#license)).
|
|
373
352
|
|
|
353
|
+
## Contributing
|
|
354
|
+
|
|
355
|
+
Issues are welcome. Run checks and tests before submitting:
|
|
356
|
+
|
|
357
|
+
```sh
|
|
358
|
+
bun run devcheck
|
|
359
|
+
bun run test
|
|
360
|
+
```
|
|
361
|
+
|
|
374
362
|
## License
|
|
375
363
|
|
|
376
364
|
Apache-2.0 — see [LICENSE](LICENSE) for details.
|
|
@@ -0,0 +1,29 @@
|
|
|
1
|
+
---
|
|
2
|
+
summary: "Adopts @cyanheads/mcp-ts-core 0.13.2, declaring stateless HTTP session mode in source, plus dependency and skills-tree maintenance."
|
|
3
|
+
breaking: false
|
|
4
|
+
security: false
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# 0.8.1 — 2026-09-16
|
|
8
|
+
|
|
9
|
+
## Changed
|
|
10
|
+
|
|
11
|
+
- **Session mode** — the server declares `stateless` HTTP session mode in source (`createApp({ sessionMode: 'stateless' })`). With `MCP_SESSION_MODE` unset, an HTTP run now serves `stateless` — previously the schema default `auto` resolved to `stateful`; an explicit `MCP_SESSION_MODE` value still overrides it. The Docker image already set `stateless`.
|
|
12
|
+
- **`/.well-known/mcp.json`** — the server card now publishes the resolved session mode under the `_meta` key `io.github.cyanheads.mcp-ts-core/sessionMode`.
|
|
13
|
+
- **Shutdown** — `SIGTERM` and `SIGINT` now end the process once shutdown settles.
|
|
14
|
+
- **`skills/` → `framework-skills/`** — the framework's development skills moved, so a plugin install no longer hands them to the installing agent.
|
|
15
|
+
|
|
16
|
+
## Fixed
|
|
17
|
+
|
|
18
|
+
- **Landing page** — the curl snippet now sends a handshake `initialize` accepts.
|
|
19
|
+
- **`.env.example`** — session and host comments corrected (`MCP_HTTP_HOST` default `127.0.0.1`); README gains an `MCP_SESSION_MODE` row.
|
|
20
|
+
|
|
21
|
+
## Dependencies
|
|
22
|
+
|
|
23
|
+
- `@cyanheads/mcp-ts-core` ^0.12.7 → ^0.13.2
|
|
24
|
+
- Bun engines floor `>=1.3.0` → `>=1.4.0`
|
|
25
|
+
- `zod` ^4.5.4 → ^4.6.4
|
|
26
|
+
- `@biomejs/biome` 2.5.12 → 2.5.13
|
|
27
|
+
- `@types/node` 26.4.1 → 26.5.1
|
|
28
|
+
- `ignore` ^7.0.8 → ^7.0.9
|
|
29
|
+
- `tsc-alias` ^1.9.4 → ^1.9.5
|
|
@@ -0,0 +1,24 @@
|
|
|
1
|
+
---
|
|
2
|
+
summary: "Adopts mcp-ts-core 0.13.6's argument-rejection recovery hints and pre-validation aliasing; rewrites server instructions for workflow chaining."
|
|
3
|
+
breaking: false
|
|
4
|
+
security: false
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# 0.8.2 — 2026-09-21
|
|
8
|
+
|
|
9
|
+
## Changed
|
|
10
|
+
|
|
11
|
+
- **Server instructions** — rewritten as three sentences addressed to the calling agent: where a workflow starts, what a PDB ID chains into, and how a running Foldseek search or structural alignment resumes.
|
|
12
|
+
- **Argument-rejection recovery** — an unknown or mistyped tool argument now carries `data.reason: "invalid_arguments"` and a `Recovery:` hint naming the accepted keys or expected type; an omitted required enum field reads as missing rather than as a wrong choice.
|
|
13
|
+
- **Argument pre-validation** — a differently-cased key naming exactly one declared parameter is rewritten (`pdbId` → `pdb_id`), and a JSON-stringified array argument is repaired and re-parsed once; an undeclared key is still rejected.
|
|
14
|
+
- **Tool-error text** — closes with `(reason <reason>)`, adding `· retryable` or `· not retryable` when the reason declares it.
|
|
15
|
+
- **Tool/field descriptions** — five definitions (`protein_analyze_collection`, `protein_compare_structures`, `protein_get_annotations`, `protein_search_structures`, `protein_track_ligands`) converted from `+`-joined strings to single string literals; advertised `tools/list` text is unchanged ([#38](https://github.com/cyanheads/protein-mcp-server/issues/38)).
|
|
16
|
+
- **Repo hygiene** — framework skills, the definition linter (config-driven truncation allowlist via `devcheck.config.json`), issue forms, and devcheck synced to the current template; adds `.github/workflows/codeql.yml`.
|
|
17
|
+
|
|
18
|
+
## Dependencies
|
|
19
|
+
|
|
20
|
+
- `@cyanheads/mcp-ts-core` ^0.13.2 → ^0.13.6
|
|
21
|
+
- `zod` ^4.6.4 → ^4.6.5
|
|
22
|
+
- `@biomejs/biome` 2.5.13 → 2.5.14
|
|
23
|
+
- `@types/node` 26.5.1 → 26.6.2
|
|
24
|
+
- `vitest` ^5.0.0 → ^5.0.1
|
package/changelog/template.md
CHANGED
|
@@ -6,8 +6,8 @@
|
|
|
6
6
|
|
|
7
7
|
# Required. One-line GitHub Release-style headline. 350 character cap — a
|
|
8
8
|
# ceiling, not a target. Default short and scannable. Don't pad, don't stitch
|
|
9
|
-
# unrelated changes with commas/semicolons into an inventory — pick the
|
|
10
|
-
# headline
|
|
9
|
+
# unrelated changes with commas/semicolons into an inventory — pick the one
|
|
10
|
+
# headline the release is about. Quotes required: unquoted YAML treats
|
|
11
11
|
# `: ` inside the value as a key separator and fails GitHub's strict parser.
|
|
12
12
|
summary: ""
|
|
13
13
|
|
|
@@ -117,30 +117,13 @@ security: false
|
|
|
117
117
|
in that unrelated item's metadata.
|
|
118
118
|
|
|
119
119
|
TAG ANNOTATIONS — the annotated tag body renders as the GitHub Release body
|
|
120
|
-
via `gh release create --notes-from-tag`.
|
|
121
|
-
|
|
122
|
-
|
|
123
|
-
|
|
124
|
-
|
|
125
|
-
|
|
126
|
-
|
|
127
|
-
Dependency bumps: ← section header
|
|
128
|
-
← blank line
|
|
129
|
-
- `@cyanheads/mcp-ts-core` ^0.9.1 → ^0.9.6 ← bullet
|
|
130
|
-
← blank line
|
|
131
|
-
Changed: ← only sections with entries
|
|
132
|
-
← blank line
|
|
133
|
-
- `format()` output includes `query` in text mode
|
|
134
|
-
← blank line
|
|
135
|
-
Added:
|
|
136
|
-
← blank line
|
|
137
|
-
- `manifest.json` scaffolded for MCPB bundle support
|
|
138
|
-
- Install badges (Claude Desktop, Cursor, VS Code)
|
|
139
|
-
← blank line
|
|
140
|
-
<N> tests pass; `bun run devcheck` clean. ← footer
|
|
141
|
-
|
|
142
|
-
Never a flat comma-separated string. Always structured markdown with
|
|
143
|
-
sections. The tag must scan well as a rendered GitHub Release page.
|
|
120
|
+
via `gh release create --notes-from-tag`. It is a condensed digest of this
|
|
121
|
+
entry, never a copy, and its format is owned by the `release-and-publish`
|
|
122
|
+
skill (step 4, "Create the annotated tag"): a short subject line without the
|
|
123
|
+
version, flat headline bullets — no Keep-a-Changelog section headers, no
|
|
124
|
+
gates line — at most one deps line, issue backlinks, and the changelog link
|
|
125
|
+
last. In release-PR mode the `git-wrapup` skill authors those bullets as the
|
|
126
|
+
PR body's `## Changes` and the tag copies them.
|
|
144
127
|
-->
|
|
145
128
|
|
|
146
129
|
## Added
|
package/dist/index.js
CHANGED
|
@@ -31,6 +31,8 @@ await createApp({
|
|
|
31
31
|
prompts: [],
|
|
32
32
|
// Public, keyless data server — serve the full inventory to unauthenticated callers.
|
|
33
33
|
landing: { requireAuth: false },
|
|
34
|
+
// No tool gates on ctx.requestInput, so HTTP serving needs no live session.
|
|
35
|
+
sessionMode: 'stateless',
|
|
34
36
|
/**
|
|
35
37
|
* Cache hints for protocol revision 2026-07-28. Every listing is static per
|
|
36
38
|
* build and identical for every caller — no auth-gated definitions, no
|
|
@@ -46,7 +48,7 @@ await createApp({
|
|
|
46
48
|
'server/discover': { ttlMs: 3_600_000, cacheScope: 'public' },
|
|
47
49
|
'resources/read': { ttlMs: 3_600_000, cacheScope: 'public' },
|
|
48
50
|
},
|
|
49
|
-
instructions: '
|
|
51
|
+
instructions: 'Find structures with protein_search_structures, then pass the returned IDs to protein_get_structure for metadata and coordinate URLs, or pass UniProt accessions there for AlphaFold predictions and the best available model. A PDB ID also chains into protein_get_annotations, protein_track_ligands, protein_find_similar, and protein_compare_structures for annotations, binding sites, homologs, and structural alignment, while protein_analyze_collection profiles the whole PDB without pulling rows. A Foldseek search or structural alignment still running when the poll budget elapses returns status "computing" with a ticket (protein_find_similar) or a job UUID (protein_compare_structures); re-call with it to resume that job rather than resubmitting.',
|
|
50
52
|
setup(core) {
|
|
51
53
|
const serverConfig = getServerConfig();
|
|
52
54
|
initRcsbService(core.config, core.storage, serverConfig);
|
package/dist/index.js.map
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":";AACA;;;;;GAKG;AAEH,OAAO,EAAE,SAAS,EAAE,MAAM,wBAAwB,CAAC;AACnD,OAAO,EAAE,eAAe,EAAE,MAAM,2BAA2B,CAAC;AAC5D,OAAO,EAAE,iBAAiB,EAAE,kBAAkB,EAAE,MAAM,6CAA6C,CAAC;AACpG,OAAO,EACL,iBAAiB,EACjB,iBAAiB,EACjB,WAAW,EACX,cAAc,EACd,YAAY,EACZ,gBAAgB,EAChB,YAAY,GACb,MAAM,yCAAyC,CAAC;AACjD,OAAO,EAAE,oBAAoB,EAAE,MAAM,2CAA2C,CAAC;AACjF,OAAO,EAAE,oBAAoB,EAAE,MAAM,2CAA2C,CAAC;AACjF,OAAO,EAAE,kBAAkB,EAAE,MAAM,uCAAuC,CAAC;AAC3E,OAAO,EAAE,mBAAmB,EAAE,MAAM,yCAAyC,CAAC;AAC9E,OAAO,EAAE,eAAe,EAAE,MAAM,iCAAiC,CAAC;AAClE,OAAO,EAAE,kBAAkB,EAAE,MAAM,uCAAuC,CAAC;AAE3E,MAAM,SAAS,CAAC;IACd,IAAI,EAAE,oBAAoB;IAC1B,KAAK,EAAE,oBAAoB;IAC3B,KAAK,EAAE;QACL,gBAAgB;QAChB,YAAY;QACZ,WAAW;QACX,YAAY;QACZ,iBAAiB;QACjB,iBAAiB;QACjB,cAAc;KACf;IACD,SAAS,EAAE,CAAC,kBAAkB,EAAE,iBAAiB,CAAC;IAClD,OAAO,EAAE,EAAE;IACX,qFAAqF;IACrF,OAAO,EAAE,EAAE,WAAW,EAAE,KAAK,EAAE;IAC/B;;;;;;;OAOG;IACH,UAAU,EAAE;QACV,YAAY,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;QACxD,gBAAgB,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;QAC5D,0BAA0B,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;QACtE,iBAAiB,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;QAC7D,gBAAgB,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;KAC7D;IACD,YAAY,EACV
|
|
1
|
+
{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":";AACA;;;;;GAKG;AAEH,OAAO,EAAE,SAAS,EAAE,MAAM,wBAAwB,CAAC;AACnD,OAAO,EAAE,eAAe,EAAE,MAAM,2BAA2B,CAAC;AAC5D,OAAO,EAAE,iBAAiB,EAAE,kBAAkB,EAAE,MAAM,6CAA6C,CAAC;AACpG,OAAO,EACL,iBAAiB,EACjB,iBAAiB,EACjB,WAAW,EACX,cAAc,EACd,YAAY,EACZ,gBAAgB,EAChB,YAAY,GACb,MAAM,yCAAyC,CAAC;AACjD,OAAO,EAAE,oBAAoB,EAAE,MAAM,2CAA2C,CAAC;AACjF,OAAO,EAAE,oBAAoB,EAAE,MAAM,2CAA2C,CAAC;AACjF,OAAO,EAAE,kBAAkB,EAAE,MAAM,uCAAuC,CAAC;AAC3E,OAAO,EAAE,mBAAmB,EAAE,MAAM,yCAAyC,CAAC;AAC9E,OAAO,EAAE,eAAe,EAAE,MAAM,iCAAiC,CAAC;AAClE,OAAO,EAAE,kBAAkB,EAAE,MAAM,uCAAuC,CAAC;AAE3E,MAAM,SAAS,CAAC;IACd,IAAI,EAAE,oBAAoB;IAC1B,KAAK,EAAE,oBAAoB;IAC3B,KAAK,EAAE;QACL,gBAAgB;QAChB,YAAY;QACZ,WAAW;QACX,YAAY;QACZ,iBAAiB;QACjB,iBAAiB;QACjB,cAAc;KACf;IACD,SAAS,EAAE,CAAC,kBAAkB,EAAE,iBAAiB,CAAC;IAClD,OAAO,EAAE,EAAE;IACX,qFAAqF;IACrF,OAAO,EAAE,EAAE,WAAW,EAAE,KAAK,EAAE;IAC/B,4EAA4E;IAC5E,WAAW,EAAE,WAAW;IACxB;;;;;;;OAOG;IACH,UAAU,EAAE;QACV,YAAY,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;QACxD,gBAAgB,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;QAC5D,0BAA0B,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;QACtE,iBAAiB,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;QAC7D,gBAAgB,EAAE,EAAE,KAAK,EAAE,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE;KAC7D;IACD,YAAY,EACV,+uBAA+uB;IACjvB,KAAK,CAAC,IAAI;QACR,MAAM,YAAY,GAAG,eAAe,EAAE,CAAC;QACvC,eAAe,CAAC,IAAI,CAAC,MAAM,EAAE,IAAI,CAAC,OAAO,EAAE,YAAY,CAAC,CAAC;QACzD,oBAAoB,CAAC,IAAI,CAAC,MAAM,EAAE,IAAI,CAAC,OAAO,EAAE,YAAY,CAAC,CAAC;QAC9D,kBAAkB,CAAC,IAAI,CAAC,MAAM,EAAE,IAAI,CAAC,OAAO,EAAE,YAAY,CAAC,CAAC;QAC5D,kBAAkB,CAAC,IAAI,CAAC,MAAM,EAAE,IAAI,CAAC,OAAO,EAAE,YAAY,CAAC,CAAC;QAC5D,oBAAoB,CAAC,IAAI,CAAC,MAAM,EAAE,IAAI,CAAC,OAAO,EAAE,YAAY,CAAC,CAAC;QAC9D,mBAAmB,CAAC,IAAI,CAAC,MAAM,EAAE,IAAI,CAAC,OAAO,EAAE,YAAY,CAAC,CAAC;IAC/D,CAAC;CACF,CAAC,CAAC"}
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"analyze-collection.tool.d.ts","sourceRoot":"","sources":["../../../../src/mcp-server/tools/definitions/analyze-collection.tool.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AAEH,OAAO,EAAQ,CAAC,EAAE,MAAM,wBAAwB,CAAC;AACjD,OAAO,EAAE,gBAAgB,EAAE,MAAM,+BAA+B,CAAC;AAuCjE,eAAO,MAAM,iBAAiB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
|
|
1
|
+
{"version":3,"file":"analyze-collection.tool.d.ts","sourceRoot":"","sources":["../../../../src/mcp-server/tools/definitions/analyze-collection.tool.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AAEH,OAAO,EAAQ,CAAC,EAAE,MAAM,wBAAwB,CAAC;AACjD,OAAO,EAAE,gBAAgB,EAAE,MAAM,+BAA+B,CAAC;AAuCjE,eAAO,MAAM,iBAAiB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;EA8L5B,CAAC"}
|
|
@@ -31,13 +31,7 @@ const ZERO_MATCH_NOTICE = {
|
|
|
31
31
|
};
|
|
32
32
|
export const analyzeCollection = tool('protein_analyze_collection', {
|
|
33
33
|
title: 'protein-mcp-server: analyze collection',
|
|
34
|
-
description: 'Profile the PDB into distributions and trends over an optional scoping query: counts by method, '
|
|
35
|
-
'organism, or polymer composition; resolution and molecular-weight histograms; release-year timelines; ' +
|
|
36
|
-
'and multidimensional cross-tabs (e.g. method × release_year). Aggregation runs server-side at RCSB — ' +
|
|
37
|
-
'one call returns compact buckets, no row pull. Pass one group_by dimension for a single breakdown, or ' +
|
|
38
|
-
'two distinct dimensions for a cross-tab (the first nests the second). bucket_limit caps each dimension ' +
|
|
39
|
-
'level separately rather than the response, so a cross-tab returns up to that many nested buckets under ' +
|
|
40
|
-
'each of its capped parent buckets; bucketsReturned reports the realized total.',
|
|
34
|
+
description: 'Profile the PDB into distributions and trends over an optional scoping query: counts by method, organism, or polymer composition; resolution and molecular-weight histograms; release-year timelines; and multidimensional cross-tabs (e.g. method × release_year). Aggregation runs server-side at RCSB — one call returns compact buckets, no row pull. Pass one group_by dimension for a single breakdown, or two distinct dimensions for a cross-tab (the first nests the second). bucket_limit caps each dimension level separately rather than the response, so a cross-tab returns up to that many nested buckets under each of its capped parent buckets; bucketsReturned reports the realized total.',
|
|
41
35
|
annotations: { readOnlyHint: true, openWorldHint: true },
|
|
42
36
|
errors: [
|
|
43
37
|
{
|
|
@@ -73,8 +67,7 @@ export const analyzeCollection = tool('protein_analyze_collection', {
|
|
|
73
67
|
content_type: z
|
|
74
68
|
.enum(['experimental', 'predicted', 'all'])
|
|
75
69
|
.default('experimental')
|
|
76
|
-
.describe('Which structure universe to profile. Default experimental. Computed models carry no '
|
|
77
|
-
'experimental metadata, so method and resolution return nothing under "predicted".'),
|
|
70
|
+
.describe('Which structure universe to profile. Default experimental. Computed models carry no experimental metadata, so method and resolution return nothing under "predicted".'),
|
|
78
71
|
interval: z
|
|
79
72
|
.union([
|
|
80
73
|
// Coerce the numeric arm: many clients stringify tool args, and "0.5" must
|
|
@@ -99,10 +92,7 @@ export const analyzeCollection = tool('protein_analyze_collection', {
|
|
|
99
92
|
.min(1)
|
|
100
93
|
.max(500)
|
|
101
94
|
.optional()
|
|
102
|
-
.describe('Max buckets per dimension level, not per response. A cross-tab applies the cap separately to the
|
|
103
|
-
'parent dimension and to the nested child inside each parent bucket, so up to ' +
|
|
104
|
-
'bucket_limit × (1 + bucket_limit) buckets can come back — 2550 at the default 50. The realized ' +
|
|
105
|
-
'count comes back as bucketsReturned. Defaults to the server PROTEIN_FACET_BUCKET_CAP.'),
|
|
95
|
+
.describe('Max buckets per dimension level, not per response. A cross-tab applies the cap separately to the parent dimension and to the nested child inside each parent bucket, so up to bucket_limit × (1 + bucket_limit) buckets can come back — 2550 at the default 50. The realized count comes back as bucketsReturned. Defaults to the server PROTEIN_FACET_BUCKET_CAP.'),
|
|
106
96
|
}),
|
|
107
97
|
output: z.object({
|
|
108
98
|
total: z.number().describe('Total entries in the scoped collection.'),
|
|
@@ -120,9 +110,7 @@ export const analyzeCollection = tool('protein_analyze_collection', {
|
|
|
120
110
|
.describe('True when at least one dimension position — the top-level dimension or a nested cross-tab child — had more buckets than the applied cap. Which positions, and by how much, is named in notice.'),
|
|
121
111
|
bucketsReturned: z
|
|
122
112
|
.number()
|
|
123
|
-
.describe('Buckets in this response, summed over every dimension level: the top-level buckets plus, for a '
|
|
124
|
-
'cross-tab, the nested child buckets under each of them. Since bucket_limit caps each level ' +
|
|
125
|
-
'separately, this is the size those caps actually produced — always present, cross-tab or not.'),
|
|
113
|
+
.describe('Buckets in this response, summed over every dimension level: the top-level buckets plus, for a cross-tab, the nested child buckets under each of them. Since bucket_limit caps each level separately, this is the size those caps actually produced — always present, cross-tab or not.'),
|
|
126
114
|
},
|
|
127
115
|
async handler(input, ctx) {
|
|
128
116
|
const cfg = getServerConfig();
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"analyze-collection.tool.js","sourceRoot":"","sources":["../../../../src/mcp-server/tools/definitions/analyze-collection.tool.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AAEH,OAAO,EAAE,IAAI,EAAE,CAAC,EAAE,MAAM,wBAAwB,CAAC;AACjD,OAAO,EAAE,gBAAgB,EAAE,MAAM,+BAA+B,CAAC;AACjE,OAAO,EAAE,eAAe,EAAE,MAAM,2BAA2B,CAAC;AAC5D,OAAO,EACL,cAAc,EACd,qBAAqB,EAErB,wBAAwB,EACxB,cAAc,GACf,MAAM,2BAA2B,CAAC;AACnC,OAAO,EAAE,cAAc,EAAE,MAAM,iCAAiC,CAAC;AACjE,OAAO,EACL,mBAAmB,EACnB,YAAY,EACZ,eAAe,EACf,oBAAoB,EACpB,YAAY,EACZ,aAAa,EACb,iBAAiB,GAClB,MAAM,eAAe,CAAC;AAEvB;;;;;GAKG;AACH,MAAM,4BAA4B,GAAG,IAAI,GAAG,CAAqB,CAAC,QAAQ,EAAE,YAAY,CAAC,CAAC,CAAC;AAE3F;;;;GAIG;AACH,MAAM,iBAAiB,GAAG;IACxB,YAAY,EAAE,wOAAwO;IACtP,SAAS,EAAE,yOAAyO;IACpP,GAAG,EAAE,mNAAmN;CAC1J,CAAC;AAEjE,MAAM,CAAC,MAAM,iBAAiB,GAAG,IAAI,CAAC,4BAA4B,EAAE;IAClE,KAAK,EAAE,wCAAwC;IAC/C,WAAW,EACT
|
|
1
|
+
{"version":3,"file":"analyze-collection.tool.js","sourceRoot":"","sources":["../../../../src/mcp-server/tools/definitions/analyze-collection.tool.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AAEH,OAAO,EAAE,IAAI,EAAE,CAAC,EAAE,MAAM,wBAAwB,CAAC;AACjD,OAAO,EAAE,gBAAgB,EAAE,MAAM,+BAA+B,CAAC;AACjE,OAAO,EAAE,eAAe,EAAE,MAAM,2BAA2B,CAAC;AAC5D,OAAO,EACL,cAAc,EACd,qBAAqB,EAErB,wBAAwB,EACxB,cAAc,GACf,MAAM,2BAA2B,CAAC;AACnC,OAAO,EAAE,cAAc,EAAE,MAAM,iCAAiC,CAAC;AACjE,OAAO,EACL,mBAAmB,EACnB,YAAY,EACZ,eAAe,EACf,oBAAoB,EACpB,YAAY,EACZ,aAAa,EACb,iBAAiB,GAClB,MAAM,eAAe,CAAC;AAEvB;;;;;GAKG;AACH,MAAM,4BAA4B,GAAG,IAAI,GAAG,CAAqB,CAAC,QAAQ,EAAE,YAAY,CAAC,CAAC,CAAC;AAE3F;;;;GAIG;AACH,MAAM,iBAAiB,GAAG;IACxB,YAAY,EAAE,wOAAwO;IACtP,SAAS,EAAE,yOAAyO;IACpP,GAAG,EAAE,mNAAmN;CAC1J,CAAC;AAEjE,MAAM,CAAC,MAAM,iBAAiB,GAAG,IAAI,CAAC,4BAA4B,EAAE;IAClE,KAAK,EAAE,wCAAwC;IAC/C,WAAW,EACT,+qBAA+qB;IACjrB,WAAW,EAAE,EAAE,YAAY,EAAE,IAAI,EAAE,aAAa,EAAE,IAAI,EAAE;IAExD,MAAM,EAAE;QACN;YACE,MAAM,EAAE,yBAAyB;YACjC,IAAI,EAAE,gBAAgB,CAAC,aAAa;YACpC,IAAI,EAAE,6HAA6H;YACnI,QAAQ,EAAE,4KAA4K;SACvL;QACD;YACE,MAAM,EAAE,qBAAqB;YAC7B,IAAI,EAAE,gBAAgB,CAAC,aAAa;YACpC,IAAI,EAAE,qFAAqF;YAC3F,QAAQ,EACN,8GAA8G;SACjH;KACF;IAED,KAAK,EAAE,CAAC,CAAC,MAAM,CAAC;QACd,QAAQ,EAAE,CAAC;aACR,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,qBAAqB,CAAC,CAAC;aACpC,GAAG,CAAC,CAAC,CAAC;aACN,GAAG,CAAC,CAAC,CAAC;aACN,QAAQ,CACP,wIAAwI,CACzI;QACH,KAAK,EAAE,CAAC;aACL,MAAM,EAAE;aACR,QAAQ,EAAE;aACV,QAAQ,CAAC,0EAA0E,CAAC;QACvF,QAAQ,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAAC,iCAAiC,CAAC;QAC3E,MAAM,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAAC,qCAAqC,CAAC;QAC7E,cAAc,EAAE,CAAC,CAAC,MAAM;aACrB,MAAM,EAAE;aACR,QAAQ,EAAE;aACV,QAAQ,EAAE;aACV,QAAQ,CAAC,wCAAwC,CAAC;QACrD,YAAY,EAAE,CAAC;aACZ,IAAI,CAAC,CAAC,cAAc,EAAE,WAAW,EAAE,KAAK,CAAC,CAAC;aAC1C,OAAO,CAAC,cAAc,CAAC;aACvB,QAAQ,CACP,uKAAuK,CACxK;QACH,QAAQ,EAAE,CAAC;aACR,KAAK,CACJ;YACE,2EAA2E;YAC3E,0EAA0E;YAC1E,+EAA+E;YAC/E,CAAC,CAAC,MAAM;iBACL,MAAM,EAAE;iBACR,QAAQ,EAAE;iBACV,QAAQ,CAAC,8DAA8D,CAAC;YAC3E,2EAA2E;YAC3E,oEAAoE;YACpE,CAAC,CAAC,IAAI,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,QAAQ,CAAC,uDAAuD,CAAC;SACnF;QACD,4EAA4E;QAC5E,wDAAwD;QACxD,EAAE,KAAK,EAAE,2EAA2E,EAAE,CACvF;aACA,QAAQ,EAAE;aACV,QAAQ,CACP,8fAA8f,CAC/f;QACH,YAAY,EAAE,CAAC,CAAC,MAAM;aACnB,MAAM,EAAE;aACR,GAAG,EAAE;aACL,GAAG,CAAC,CAAC,CAAC;aACN,GAAG,CAAC,GAAG,CAAC;aACR,QAAQ,EAAE;aACV,QAAQ,CACP,oWAAoW,CACrW;KACJ,CAAC;IAEF,MAAM,EAAE,CAAC,CAAC,MAAM,CAAC;QACf,KAAK,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,CAAC,yCAAyC,CAAC;QACrE,MAAM,EAAE,CAAC,CAAC,KAAK,CAAC,oBAAoB,CAAC,CAAC,QAAQ,CAAC,6BAA6B,CAAC;KAC9E,CAAC;IAEF,UAAU,EAAE;QACV,KAAK,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAAC,+CAA+C,CAAC;QACtF,MAAM,EAAE,CAAC;aACN,MAAM,EAAE;aACR,QAAQ,EAAE;aACV,QAAQ,CACP,yPAAyP,CAC1P;QACH,SAAS,EAAE,CAAC;aACT,OAAO,EAAE;aACT,QAAQ,EAAE;aACV,QAAQ,CACP,gMAAgM,CACjM;QACH,eAAe,EAAE,CAAC;aACf,MAAM,EAAE;aACR,QAAQ,CACP,yRAAyR,CAC1R;KACJ;IAED,KAAK,CAAC,OAAO,CAAC,KAAK,EAAE,GAAG;QACtB,MAAM,GAAG,GAAG,eAAe,EAAE,CAAC;QAC9B,MAAM,IAAI,GAAG,cAAc,EAAE,CAAC;QAC9B,MAAM,GAAG,GAAG,KAAK,CAAC,YAAY,IAAI,GAAG,CAAC,cAAc,CAAC;QACrD,8EAA8E;QAC9E,kFAAkF;QAClF,MAAM,CAAC,OAAO,EAAE,SAAS,CAAC,GAAG,KAAK,CAAC,QAAqD,CAAC;QACzF,+EAA+E;QAC/E,6EAA6E;QAC7E,uEAAuE;QACvE,MAAM,SAAS,GAAG,KAAK,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC;QAC3E,IAAI,SAAS;YACX,MAAM,GAAG,CAAC,IAAI,CACZ,qBAAqB,EACrB,mBAAmB,SAAS,qDAAqD,EACjF,EAAE,GAAG,GAAG,CAAC,WAAW,CAAC,qBAAqB,CAAC,EAAE,CAC9C,CAAC;QACJ,+EAA+E;QAC/E,kFAAkF;QAClF,IAAI,KAAK,CAAC,QAAQ,KAAK,SAAS,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,QAAQ,EAAE,OAAO,EAAE,SAAS,CAAC;YACrF,MAAM,GAAG,CAAC,IAAI,CACZ,yBAAyB,EACzB,YAAY,KAAK,CAAC,QAAQ,sBAAsB,KAAK,CAAC,QAAQ,CAAC,IAAI,CAAC,MAAM,CAAC,UAAU,wBAAwB,CAAC,IAAI,CAAC,IAAI,CAAC,gHAAgH,EACxO,EAAE,GAAG,GAAG,CAAC,WAAW,CAAC,yBAAyB,CAAC,EAAE,CAClD,CAAC;QACJ,MAAM,IAAI,GAAG,cAAc,CAAC,OAAO,EAAE,KAAK,CAAC,QAAQ,EAAE,SAAS,CAAC,CAAC;QAEhE,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,GAAG,MAAM,IAAI,CAAC,aAAa,CAChD;YACE,GAAG,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,KAAK,CAAC,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;YAC7C,GAAG,CAAC,KAAK,CAAC,QAAQ,CAAC,CAAC,CAAC,EAAE,QAAQ,EAAE,KAAK,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;YACvD,GAAG,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;YACjD,GAAG,CAAC,OAAO,KAAK,CAAC,cAAc,KAAK,QAAQ;gBAC1C,CAAC,CAAC,EAAE,aAAa,EAAE,KAAK,CAAC,cAAc,EAAE;gBACzC,CAAC,CAAC,EAAE,CAAC;YACP,WAAW,EAAE,mBAAmB,CAAC,KAAK,CAAC,YAAY,CAAC;SACrD,EACD,CAAC,IAAI,CAAC,EACN,GAAG,CACJ,CAAC;QAEF,MAAM,GAAG,GAAG,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,aAAa,CAAC,CAAC,EAAE,GAAG,EAAE,KAAK,CAAC,CAAC,CAAC;QAE5D,+EAA+E;QAC/E,6EAA6E;QAC7E,GAAG,CAAC,MAAM,CAAC,EAAE,eAAe,EAAE,YAAY,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QAEnD,6EAA6E;QAC7E,4EAA4E;QAC5E,4EAA4E;QAC5E,MAAM,OAAO,GAAa,EAAE,CAAC;QAC7B,iFAAiF;QACjF,0EAA0E;QAC1E,MAAM,WAAW,GAAG,iBAAiB,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC;QAChD,IAAI,WAAW,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YAC3B,GAAG,CAAC,MAAM,CAAC,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;YAChC,OAAO,CAAC,IAAI,CAAC,GAAG,WAAW,CAAC,CAAC;QAC/B,CAAC;QACD,IAAI,KAAK,KAAK,CAAC,EAAE,CAAC;YAChB,yEAAyE;YACzE,6EAA6E;YAC7E,OAAO,CAAC,IAAI,CAAC,iBAAiB,CAAC,KAAK,CAAC,YAAY,CAAC,CAAC,CAAC;QACtD,CAAC;aAAM,IAAI,KAAK,CAAC,YAAY,KAAK,WAAW,EAAE,CAAC;YAC9C,MAAM,KAAK,GAAG,KAAK,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,4BAA4B,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;YAChF,IAAI,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACrB,OAAO,CAAC,IAAI,CACV,GAAG,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,IAAI,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,8NAA8N,CACxR,CAAC;YACJ,CAAC;QACH,CAAC;QACD,OAAO,CAAC,IAAI,CAAC,GAAG,eAAe,CAAC,GAAG,EAAE,KAAK,CAAC,CAAC,CAAC;QAC7C,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC;YAAE,GAAG,CAAC,MAAM,CAAC,MAAM,CAAC,OAAO,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC;QAE7D,MAAM,SAAS,GAAG,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,QAAQ,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;QAC9E,IAAI,SAAS,CAAC,MAAM,GAAG,CAAC;YAAE,GAAG,CAAC,MAAM,CAAC,EAAE,KAAK,EAAE,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC,CAAC;QAEvE,OAAO,EAAE,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,CAAC;IAChC,CAAC;IAED,MAAM,EAAE,CAAC,MAAM,EAAE,EAAE;QACjB,MAAM,KAAK,GAAa,CAAC,2BAA2B,MAAM,CAAC,KAAK,UAAU,CAAC,CAAC;QAC5E,KAAK,CAAC,IAAI,CAAC,GAAG,YAAY,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,CAAC;QAC3C,OAAO,CAAC,EAAE,IAAI,EAAE,MAAM,EAAE,IAAI,EAAE,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;IACpD,CAAC;CACF,CAAC,CAAC"}
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"compare-structures.tool.d.ts","sourceRoot":"","sources":["../../../../src/mcp-server/tools/definitions/compare-structures.tool.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAO,EAAQ,CAAC,EAAE,MAAM,wBAAwB,CAAC;AACjD,OAAO,EAAE,gBAAgB,EAAE,MAAM,+BAA+B,CAAC;
|
|
1
|
+
{"version":3,"file":"compare-structures.tool.d.ts","sourceRoot":"","sources":["../../../../src/mcp-server/tools/definitions/compare-structures.tool.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAO,EAAQ,CAAC,EAAE,MAAM,wBAAwB,CAAC;AACjD,OAAO,EAAE,gBAAgB,EAAE,MAAM,+BAA+B,CAAC;AAqHjE,eAAO,MAAM,iBAAiB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;EAwJ5B,CAAC"}
|
|
@@ -71,10 +71,7 @@ const outputSchema = z.object({
|
|
|
71
71
|
tmScore: z
|
|
72
72
|
.number()
|
|
73
73
|
.optional()
|
|
74
|
-
.describe('TM-score (0–1; higher is more similar). Length-normalized, so it can be sensitive to '
|
|
75
|
-
'terminal length differences between the two structures — a one-residue overhang can flip ' +
|
|
76
|
-
'the greedy superposition into a worse local optimum, dropping the score sharply. Cross-check ' +
|
|
77
|
-
'rmsd and alignedResidues to spot such cases.'),
|
|
74
|
+
.describe('TM-score (0–1; higher is more similar). Length-normalized, so it can be sensitive to terminal length differences between the two structures — a one-residue overhang can flip the greedy superposition into a worse local optimum, dropping the score sharply. Cross-check rmsd and alignedResidues to spot such cases.'),
|
|
78
75
|
rmsd: z.number().optional().describe('RMSD in Å over aligned residues.'),
|
|
79
76
|
alignedResidues: z.number().optional().describe('Number of aligned residue pairs.'),
|
|
80
77
|
modeledResidues: z
|
|
@@ -98,17 +95,7 @@ const outputSchema = z.object({
|
|
|
98
95
|
});
|
|
99
96
|
export const compareStructures = tool('protein_compare_structures', {
|
|
100
97
|
title: 'protein-mcp-server: compare structures',
|
|
101
|
-
description:
|
|
102
|
-
'Comparison service (TM-align / jFATCAT). reference:"first" aligns every structure to the first; ' +
|
|
103
|
-
'reference:"all_pairs" computes the full pairwise matrix. Each pair is an independent async alignment ' +
|
|
104
|
-
'job, fanned out with a concurrency cap and per-pair partial success — a pair still computing when the ' +
|
|
105
|
-
'budget elapses returns status "computing" with its job UUID, and a failed pair degrades its row without ' +
|
|
106
|
-
"sinking the others. Re-call with a matching entry in resume[] to poll a computing pair's UUID instead " +
|
|
107
|
-
"of resubmitting. Returns TM-score, RMSD, and aligned-residue count per pair, plus each structure's " +
|
|
108
|
-
'modeled-residue count and alignment coverage. TM-score is ' +
|
|
109
|
-
'length-normalized and can shift sharply between structures that differ only by a terminal residue or ' +
|
|
110
|
-
'two — the greedy superposition can settle into a worse local optimum — so read tmScore alongside rmsd, ' +
|
|
111
|
-
'alignedResidues, modeledResidues and coverage, the columns that make such cases diagnosable.',
|
|
98
|
+
description: `Structurally align multiple structures (up to the configured batch cap) via the RCSB Structural Comparison service (TM-align / jFATCAT). reference:"first" aligns every structure to the first; reference:"all_pairs" computes the full pairwise matrix. Each pair is an independent async alignment job, fanned out with a concurrency cap and per-pair partial success — a pair still computing when the budget elapses returns status "computing" with its job UUID, and a failed pair degrades its row without sinking the others. Re-call with a matching entry in resume[] to poll a computing pair's UUID instead of resubmitting. Returns TM-score, RMSD, and aligned-residue count per pair, plus each structure's modeled-residue count and alignment coverage. TM-score is length-normalized and can shift sharply between structures that differ only by a terminal residue or two — the greedy superposition can settle into a worse local optimum — so read tmScore alongside rmsd, alignedResidues, modeledResidues and coverage, the columns that make such cases diagnosable.`,
|
|
112
99
|
annotations: { readOnlyHint: true, openWorldHint: true },
|
|
113
100
|
errors: [
|
|
114
101
|
{
|