File size: 27,655 Bytes
891fe52
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
# API Reference

Complete reference for all Polyhedra MCP tools.

**Version 2.1** includes 12 tools total (10 from v2.0 + 2 new literature review tools).

## Table of Contents

- [API Reference](#api-reference)
  - [Table of Contents](#table-of-contents)
  - [Literature Search](#literature-search)
    - [search\_papers](#search_papers)
    - [get\_paper](#get_paper)
    - [query\_similar\_papers](#query_similar_papers)
    - [index\_papers](#index_papers)
  - [Citation Management](#citation-management)
    - [add\_citation](#add_citation)
    - [get\_citations](#get_citations)
  - [File Operations](#file-operations)
    - [save\_file](#save_file)
    - [get\_context](#get_context)
  - [Project Management](#project-management)
    - [init\_project](#init_project)
    - [get\_project\_status](#get_project_status)
  - [Literature Review Generation (v2.1)](#literature-review-generation-v21)
    - [generate\_literature\_review](#generate_literature_review)
    - [estimate\_review\_cost](#estimate_review_cost)
  - [Error Handling](#error-handling)
    - [Error Response Format](#error-response-format)
    - [Common Error Types](#common-error-types)
    - [Automatic Retries](#automatic-retries)
    - [Graceful Degradation](#graceful-degradation)
  - [Rate Limits \& Performance](#rate-limits--performance)
    - [Semantic Scholar API](#semantic-scholar-api)
    - [Local Operations](#local-operations)
    - [Best Practices](#best-practices)
  - [Version Compatibility](#version-compatibility)

---

## Literature Search

### search_papers

Search for academic papers using Semantic Scholar.

**Purpose**: Discover papers by keywords, authors, or topics across 200M+ publications.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `query` | string | Yes | Search query (keywords, author names, topics) |
| `year_range` | string | No | Filter by publication year (e.g., "2020-2024") |
| `limit` | integer | No | Maximum results (default: 10, max: 100) |
| `fields` | array[string] | No | Fields to include (default: title, authors, year, abstract, citationCount) |

**Returns**:

```json
{
  "papers": [
    {
      "paperId": "abc123",
      "title": "Attention Is All You Need",
      "authors": [{"name": "Ashish Vaswani"}],
      "year": 2017,
      "abstract": "The dominant sequence...",
      "citationCount": 75000,
      "url": "https://semanticscholar.org/paper/abc123"
    }
  ],
  "total": 1,
  "query": "transformers attention"
}
```

**Example Usage**:

```
Search for papers on "neural networks" published 2022-2024 with high citations
```

**Common Patterns**:

- **By topic**: `"transformers in nlp"`
- **By author**: `"author:Hinton neural networks"`
- **Recent papers**: Combine with `year_range: "2023-2024"`
- **Highly cited**: Sort results by `citationCount`

**Error Scenarios**:

- **Network error**: Returns cached results if available
- **Invalid year_range**: Falls back to all years
- **Empty results**: Suggests query refinements
- **Rate limit**: Automatically retries with backoff

**Related Tools**:
- Use `get_paper` for detailed information on specific results
- Use `index_papers` to enable semantic search on results

---

### get_paper

Get detailed information about a specific paper.

**Purpose**: Retrieve comprehensive metadata for a single paper by ID.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `paper_id` | string | Yes | Semantic Scholar paper ID (e.g., "abc123") |
| `fields` | array[string] | No | Fields to retrieve (default: all available) |

**Returns**:

```json
{
  "paperId": "abc123",
  "title": "Attention Is All You Need",
  "authors": [
    {"authorId": "456", "name": "Ashish Vaswani"}
  ],
  "year": 2017,
  "abstract": "The dominant sequence...",
  "citationCount": 75000,
  "referenceCount": 35,
  "influentialCitationCount": 12000,
  "venue": "NeurIPS",
  "url": "https://semanticscholar.org/paper/abc123",
  "openAccessPdf": {"url": "https://arxiv.org/pdf/1706.03762.pdf"},
  "fieldsOfStudy": ["Computer Science"],
  "citations": [...],
  "references": [...]
}
```

**Example Usage**:

```
Get detailed information for paper abc123
```

**Common Patterns**:

- **Get full metadata**: Use default fields
- **Get citations only**: `fields: ["citations"]`
- **Get references**: `fields: ["references"]`
- **Check PDF availability**: Look for `openAccessPdf`

**Error Scenarios**:

- **Invalid paper_id**: Returns error with suggested format
- **Paper not found**: Returns 404 with similar papers suggestion
- **Network error**: Returns cached data if available

**Related Tools**:
- Use after `search_papers` to get full details
- Use with `add_citation` to add to bibliography

---

### query_similar_papers

Find papers similar to a query using semantic search.

**Purpose**: Discover related papers through RAG-powered semantic similarity.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `query` | string | Yes | Natural language query or research question |
| `limit` | integer | No | Maximum results (default: 10) |
| `min_similarity` | float | No | Minimum similarity score 0-1 (default: 0.7) |

**Returns**:

```json
{
  "papers": [
    {
      "paperId": "abc123",
      "title": "Attention Is All You Need",
      "similarity": 0.92,
      "reason": "Discusses transformer architecture and attention mechanisms"
    }
  ],
  "query": "self-attention in neural networks"
}
```

**Example Usage**:

```
Find papers similar to "improving model interpretability with attention visualization"
```

**Common Patterns**:

- **Research questions**: `"How do transformers handle long sequences?"`
- **Topic exploration**: `"alternatives to BERT for document classification"`
- **Methodology search**: `"training strategies for large language models"`
- **Gap analysis**: Find what's missing in current literature

**Requirements**:

- Must call `index_papers` first to build search index
- Index must contain relevant papers for query

**Error Scenarios**:

- **Index not built**: Returns message to call `index_papers` first
- **No similar papers**: Returns empty list with suggestion to search more papers
- **Model not loaded**: Downloads embedding model (~400MB) automatically

**Related Tools**:
- Use `index_papers` before querying
- Use `search_papers` to collect papers to index

---

### index_papers

Build semantic search index from collected papers.

**Purpose**: Enable RAG-powered similarity search across your paper collection.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `papers` | array[object] | Yes | List of papers to index (from `search_papers`) |
| `force_rebuild` | boolean | No | Rebuild index even if exists (default: false) |

**Paper Object Schema**:

```json
{
  "paperId": "abc123",
  "title": "Paper Title",
  "abstract": "Paper abstract text...",
  "year": 2023,
  "authors": [{"name": "Author Name"}]
}
```

**Returns**:

```json
{
  "indexed_count": 25,
  "total_papers": 25,
  "index_path": "/path/to/.polyhedra/embeddings.index",
  "model": "sentence-transformers/all-MiniLM-L6-v2"
}
```

**Example Usage**:

```
Index the papers I just searched for to enable semantic search
```

**Common Patterns**:

1. **Search + Index workflow**:
   ```
   1. Search for papers on topic X
   2. Index the results
   3. Query similar papers with research questions
   ```

2. **Incremental indexing**:
   - Index papers as you discover them
   - `force_rebuild: false` appends to existing index

3. **Rebuild index**:
   - Use `force_rebuild: true` to start fresh
   - Useful when changing paper collection focus

**Performance Notes**:

- First run downloads embedding model (~400MB)
- Indexing 100 papers takes ~10-30 seconds
- Index is persisted in `.polyhedra/embeddings.index`

**Error Scenarios**:

- **Missing abstract**: Uses title only for indexing
- **Duplicate papers**: Automatically deduplicated by paperId
- **Disk full**: Returns error with space requirements

**Related Tools**:
- Use `search_papers` to collect papers to index
- Use `query_similar_papers` after indexing

---

## Citation Management

### add_citation

Add BibTeX entry to project bibliography.

**Purpose**: Maintain formatted citations for papers you reference.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `citation_key` | string | Yes | Unique BibTeX key (e.g., "vaswani2017attention") |
| `bibtex` | string | Yes | Complete BibTeX entry |

**BibTeX Format**:

```bibtex
@article{vaswani2017attention,
  title={Attention Is All You Need},
  author={Vaswani, Ashish and Shazeer, Noam and ...},
  journal={Advances in neural information processing systems},
  volume={30},
  year={2017}
}
```

**Returns**:

```json
{
  "success": true,
  "citation_key": "vaswani2017attention",
  "file": "references.bib",
  "total_citations": 42
}
```

**Example Usage**:

```
Add this BibTeX citation with key "hinton2012imagenet": @article{...}
```

**Common Patterns**:

1. **From paper details**:
   ```
   1. Get paper details with get_paper
   2. Generate BibTeX from metadata
   3. Add to bibliography
   ```

2. **Batch citations**:
   - Add multiple citations in sequence
   - Automatically creates `references.bib` if missing

3. **Key naming conventions**:
   - `author_year_keyword` (e.g., `vaswani2017attention`)
   - `firstauthor_etal_year` (e.g., `brown_etal_2020`)

**Validation**:

- Checks for valid BibTeX syntax
- Prevents duplicate citation keys
- Validates required fields (@article, @inproceedings, etc.)

**Error Scenarios**:

- **Invalid BibTeX**: Returns syntax error with line number
- **Duplicate key**: Suggests appending suffix (e.g., `_v2`)
- **File locked**: Retries with exponential backoff
- **Disk full**: Returns error with space needed

**Related Tools**:
- Use `get_citations` to view all citations
- Use `get_paper` to get BibTeX for a paper

---

### get_citations

List all citations in project bibliography.

**Purpose**: View and manage your collected citations.

**Parameters**: None

**Returns**:

```json
{
  "citations": [
    {
      "key": "vaswani2017attention",
      "type": "article",
      "title": "Attention Is All You Need",
      "authors": "Vaswani, Ashish and Shazeer, Noam and ...",
      "year": "2017",
      "venue": "Advances in neural information processing systems"
    }
  ],
  "total": 1,
  "file": "references.bib"
}
```

**Example Usage**:

```
Show me all the citations in my bibliography
```

**Common Patterns**:

1. **Review citations**:
   - Check what papers you've cited
   - Find citation keys for writing

2. **Export bibliography**:
   - Copy references.bib to document
   - Use with LaTeX `\bibliography{references}`

3. **Citation search**:
   - Find citation key by author/title
   - Check if paper already cited

**Output Format**:

- Parses BibTeX into structured JSON
- Sorted by citation key alphabetically
- Includes total count for overview

**Error Scenarios**:

- **No references.bib**: Returns empty list with suggestion to add citations
- **Corrupted file**: Returns error with validation details
- **Read permission**: Returns error with file path

**Related Tools**:
- Use `add_citation` to add new citations
- Use `save_file` to export formatted bibliography

---

## File Operations

### save_file

Write content to a project file.

**Purpose**: Create or update research documents, notes, and outputs.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `path` | string | Yes | File path relative to project root |
| `content` | string | Yes | Content to write |
| `create_dirs` | boolean | No | Create parent directories (default: true) |

**Returns**:

```json
{
  "success": true,
  "path": "literature-review/transformers.md",
  "bytes_written": 1024,
  "created": true
}
```

**Example Usage**:

```
Save this literature review to "papers/summary.md": [content]
```

**Common Patterns**:

1. **Research notes**:
   ```
   Save paper summaries to notes/
   Track ideas in ideas.md
   ```

2. **Generated content**:
   ```
   Write literature review sections
   Generate paper outlines
   Create annotated bibliographies
   ```

3. **Data organization**:
   ```
   Save paper metadata as JSON
   Export search results
   Store processed data
   ```

**Safety Features**:

- Creates parent directories automatically
- UTF-8 encoding for international characters
- Atomic writes (tmp file + rename)
- Preserves original on write failure

**Path Resolution**:

- Relative to project root
- Supports nested paths: `papers/2024/summary.md`
- Normalizes path separators (cross-platform)

**Error Scenarios**:

- **Invalid path**: Returns error with valid path format
- **Permission denied**: Returns error with required permissions
- **Disk full**: Returns error with space needed
- **Invalid UTF-8**: Returns encoding error

**Related Tools**:
- Use `get_context` to read files
- Use `get_project_status` to see project structure

---

### get_context

Read content from project files.

**Purpose**: Access research documents, notes, and data files.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `paths` | array[string] | Yes | List of file paths to read (relative to project) |

**Returns**:

```json
{
  "files": [
    {
      "path": "notes/summary.md",
      "content": "# Literature Review\n\n...",
      "size": 1024,
      "encoding": "utf-8"
    }
  ],
  "total": 1
}
```

**Example Usage**:

```
Read the files: notes/summary.md, papers/metadata.json
```

**Common Patterns**:

1. **Review research**:
   ```
   Read literature review drafts
   Check project notes
   Review paper summaries
   ```

2. **Context for generation**:
   ```
   Read existing content before updating
   Get context for writing new sections
   Check what's already documented
   ```

3. **Batch reading**:
   ```
   Read all markdown files in notes/
   Get all paper summaries
   Load configuration files
   ```

**File Types Supported**:

- Text files: `.md`, `.txt`, `.json`, `.yaml`
- Code: `.py`, `.js`, `.tex`
- Config: `.bib`, `.toml`, `.ini`
- Any UTF-8 encoded file

**Performance**:

- Reads up to 10 files in parallel
- Max file size: 10MB per file
- Returns errors for individual files without failing entire operation

**Error Scenarios**:

- **File not found**: Returns empty content with error message
- **Permission denied**: Skips file with error in response
- **Binary file**: Returns error suggesting text files only
- **Encoding error**: Tries fallback encodings (latin-1, cp1252)

**Related Tools**:
- Use `save_file` to write files
- Use `get_project_status` to see available files

---

## Project Management

### init_project

Initialize a new research project structure.

**Purpose**: Create standardized directory structure and starter files.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `project_name` | string | Yes | Name of the project (used for directory naming) |
| `base_path` | string | No | Parent directory (default: current working directory) |

**Created Structure**:

```
project_name/
β”œβ”€β”€ .polyhedra/           # Tool data (index, cache)
β”œβ”€β”€ papers/               # Paper PDFs and metadata
β”œβ”€β”€ notes/                # Research notes
β”œβ”€β”€ literature-review/    # Literature review drafts
β”œβ”€β”€ data/                 # Datasets and processed data
β”œβ”€β”€ references.bib        # BibTeX bibliography
└── README.md             # Project overview
```

**Returns**:

```json
{
  "project_name": "my-research",
  "path": "/home/user/my-research",
  "created_dirs": [
    ".polyhedra",
    "papers",
    "notes",
    "literature-review",
    "data"
  ],
  "created_files": [
    "references.bib",
    "README.md"
  ]
}
```

**Example Usage**:

```
Initialize a new research project called "transformer-survey"
```

**Common Patterns**:

1. **Start new research**:
   ```
   1. Initialize project
   2. Set project as working directory
   3. Begin searching for papers
   ```

2. **Organize existing research**:
   ```
   1. Create project structure
   2. Move existing files into structure
   3. Start using tools
   ```

**Idempotent Operation**:

- Safe to run multiple times
- Won't overwrite existing files
- Only creates missing directories
- Reports what was created vs. already existed

**README Template**:

```markdown
# [Project Name]

## Research Question

TODO: Define your research question

## Papers

Papers are organized in the `papers/` directory.

## Notes

Research notes are in the `notes/` directory.

## Citations

Bibliography is in `references.bib`.
```

**Error Scenarios**:

- **Directory exists**: Reports existing, doesn't fail
- **Permission denied**: Returns error with required permissions
- **Invalid project name**: Suggests valid naming (alphanumeric, dashes, underscores)

**Related Tools**:
- Use `get_project_status` to see project structure
- Use `save_file` and `get_context` for file operations

---

### get_project_status

Get overview of project structure and contents.

**Purpose**: Understand project organization and find files.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `path` | string | No | Project root path (default: current directory) |

**Returns**:

```json
{
  "project_path": "/home/user/my-research",
  "structure": {
    "papers": {
      "file_count": 5,
      "total_size": "25.3 MB",
      "files": ["paper1.pdf", "paper2.pdf"]
    },
    "notes": {
      "file_count": 3,
      "files": ["summary.md", "ideas.md"]
    },
    "literature-review": {
      "file_count": 1,
      "files": ["draft.md"]
    }
  },
  "citations": {
    "total": 42,
    "file": "references.bib"
  },
  "index": {
    "exists": true,
    "papers_indexed": 25,
    "last_updated": "2024-01-15T10:30:00Z"
  }
}
```

**Example Usage**:

```
Show me the status of my research project
```

**Common Patterns**:

1. **Project overview**:
   ```
   Check what files exist
   See how many papers collected
   Review citation count
   ```

2. **Before writing**:
   ```
   Check existing structure
   Find where to save new content
   See what's already documented
   ```

3. **Progress tracking**:
   ```
   Monitor paper collection growth
   Track documentation progress
   Check index status
   ```

**Status Indicators**:

- **Papers**: Count, total size, recent additions
- **Notes**: File count, last modified
- **Literature Review**: Section count, word count
- **Citations**: Total count in bibliography
- **Index**: Existence, paper count, freshness

**Error Scenarios**:

- **Not a project**: Suggests running `init_project`
- **Permission denied**: Returns error with path
- **Corrupted structure**: Reports issues found

**Related Tools**:
- Use `init_project` to create project
- Use `get_context` to read specific files
- Use `get_citations` for detailed citation info

---

## Literature Review Generation (v2.1)

### generate_literature_review

Generate structured academic literature review from paper collection.

**Purpose**: Synthesize 10-100 papers into coherent academic review with proper citations, taxonomy, critical analysis, and research gaps.

**Prerequisites**: 
- LLM API key configured (`ANTHROPIC_API_KEY` or `OPENAI_API_KEY`)
- Papers file from `search_papers` or manually created JSON

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `papers_file` | string | No | Path to papers JSON (default: "literature/papers.json") |
| `focus` | string | No | Specific focus area for review |
| `structure` | string | No | Organization: "thematic"/"chronological"/"methodological" (default: "thematic") |
| `depth` | string | No | Review depth: "brief"/"standard"/"comprehensive" (default: "standard") |
| `include_gaps` | boolean | No | Include research gaps section (default: true) |
| `output_path` | string | No | Where to save review (default: "literature/review.md") |
| `llm_model` | string | No | Model override (default: claude-3-5-sonnet-20241022) |

**Returns**:

```json
{
  "success": true,
  "saved_to": "literature/review.md",
  "metadata": {
    "paper_count": 47,
    "word_count": 3542,
    "sections": ["Overview", "Taxonomy", "Critical Analysis", "Research Gaps", "Conclusion"],
    "citations_found": 45,
    "citation_coverage": 0.96,
    "research_gaps": [
      {
        "title": "Efficiency at Scale",
        "description": "Current transformers struggle with sequences >10K tokens"
      }
    ]
  },
  "cost": {
    "input_tokens": 18234,
    "output_tokens": 5890,
    "total_tokens": 24124,
    "total_usd": 0.14
  },
  "citations_added": 47
}
```

**Review Depth Levels**:

| Depth | Words | Pages | Cost (50 papers) | Use Case |
|-------|-------|-------|------------------|----------|
| **brief** | ~650 | 2-3 | $0.08-0.12 | Quick overview, presentation prep |
| **standard** | ~2000 | 5-8 | $0.12-0.20 | Paper background section, grant proposals |
| **comprehensive** | ~2500 | 10-15 | $0.20-0.35 | Dissertation chapters, major surveys |

**Structure Types**:

- **thematic**: Groups papers by research themes and approaches (recommended for most cases)
- **chronological**: Orders papers by publication timeline (good for tracking field evolution)
- **methodological**: Organizes by research methods used (good for comparing techniques)

**Example Usage**:

```
"Generate a literature review from my papers focused on mobile deployment"
```

```
"Generate a comprehensive review with chronological structure"
```

```
"Generate brief review without research gaps, save to method/background.md"
```

**Common Patterns**:

1. **Quick Synopsis**: Use `depth="brief"` for rapid overview
2. **Focused Review**: Set `focus` parameter to narrow scope
3. **Cost-Conscious**: Use `estimate_review_cost` first
4. **Custom Location**: Set `output_path` to save anywhere in project
5. **Chronological Survey**: Use `structure="chronological"` for historical perspective

**Error Scenarios**:

| Error | Cause | Solution |
|-------|-------|----------|
| "LLM service not configured" | Missing API key | Set `ANTHROPIC_API_KEY` or `OPENAI_API_KEY` |
| "Papers file not found" | No papers.json | Run `search_papers` first or create file |
| "Papers file is empty" | Empty JSON array | Search for papers or add manually |
| "API rate limit exceeded" | Too many requests | Wait 60 seconds and retry |
| "Token limit exceeded" | Too many papers | Reduce paper count or use brief depth |

**Cost Guidance**:

- **25 papers, brief**: $0.05-0.08
- **50 papers, standard**: $0.12-0.20
- **75 papers, comprehensive**: $0.25-0.40
- **100 papers, comprehensive**: $0.35-0.60

Cost scales with:
- Number of papers (more input tokens)
- Review depth (more output tokens)
- Model choice (Claude Sonnet < GPT-4 Turbo)

**Output Format**:

Generated review includes:
- **Overview**: Field introduction and scope
- **Taxonomy**: Hierarchical organization of approaches
- **Critical Analysis**: Strengths, weaknesses, comparisons
- **Research Gaps** (if enabled): Underexplored areas and opportunities
- **Conclusion**: Summary and future directions
- **Proper Citations**: All papers cited as `[Author et al., Year]`

**Related Tools**:
- Use `search_papers` to find papers first
- Use `estimate_review_cost` to check cost before generation
- Use `add_citation` to add more citations manually
- Use `save_file` to modify and save the review

---

### estimate_review_cost

Estimate cost before generating literature review.

**Purpose**: Get accurate cost estimate without consuming API credits. Helps budget and plan expensive operations.

**Parameters**:

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `paper_count` | number | Yes | Number of papers to review |
| `depth` | string | No | Review depth (default: "standard") |
| `llm_model` | string | No | Model to use (affects pricing) |

**Returns**:

```json
{
  "estimated_input_tokens": 18000,
  "estimated_output_tokens": 2000,
  "estimated_total_tokens": 20000,
  "estimated_usd": 0.15,
  "paper_count": 50,
  "depth": "standard"
}
```

**Example Usage**:

```
"Estimate cost for reviewing 75 papers with comprehensive depth"
```

```
"How much will it cost to generate a brief review of 30 papers?"
```

**Cost Examples**:

| Papers | Depth | Model | Estimated Cost |
|--------|-------|-------|----------------|
| 25 | brief | Claude Sonnet | $0.05-0.08 |
| 50 | standard | Claude Sonnet | $0.12-0.20 |
| 50 | standard | GPT-4 Turbo | $0.30-0.40 |
| 75 | comprehensive | Claude Sonnet | $0.25-0.40 |
| 100 | comprehensive | Claude Sonnet | $0.40-0.65 |

**When to Use**:

- Before generating reviews with >50 papers
- When using comprehensive depth
- When working with budget constraints
- Before batch processing multiple reviews

**Accuracy**: Estimates are typically within Β±20% of actual cost. Actual cost depends on:
- Actual paper abstract lengths (estimated at 300 words avg)
- LLM's response length (may vary from target)
- Model pricing changes

**Related Tools**:
- Use before `generate_literature_review` for cost-conscious workflows
- Combine with `get_project_status` to track total costs

---

## Error Handling

All tools follow consistent error handling patterns:

### Error Response Format

```json
{
  "error": true,
  "message": "Human-readable error description",
  "error_type": "NetworkError",
  "details": {
    "path": "/path/to/file",
    "suggestion": "Try checking your internet connection"
  }
}
```

### Common Error Types

| Error Type | Description | Recovery |
|------------|-------------|----------|
| `NetworkError` | API or connection issues | Retries automatically, uses cache |
| `FileNotFoundError` | File doesn't exist | Returns suggestion for valid path |
| `ValidationError` | Invalid input parameters | Returns validation details |
| `RateLimitError` | API rate limit exceeded | Waits and retries automatically |
| `IndexNotBuiltError` | Semantic index missing | Suggests calling `index_papers` |

### Automatic Retries

- **Network errors**: 3 retries with exponential backoff
- **Rate limits**: Waits for reset time, then retries
- **Transient errors**: Retries with jitter

### Graceful Degradation

- **Offline mode**: Uses cached data when available
- **Partial results**: Returns successful operations even if some fail
- **Fallbacks**: Switches to alternative methods when primary fails

See [Error Handling Guide](ERROR_HANDLING.md) for detailed troubleshooting.

---

## Rate Limits & Performance

### Semantic Scholar API

- **Rate limit**: 100 requests/5 minutes
- **Burst**: 10 requests/second
- **Handling**: Automatic backoff and retry

### Local Operations

- **File reads**: No limit, cached in memory
- **File writes**: No limit, atomic operations
- **Indexing**: ~100 papers/minute (after model download)

### Best Practices

1. **Batch operations** when possible
2. **Reuse index** instead of rebuilding
3. **Cache search results** for repeated queries
4. **Use specific queries** to reduce API calls

---

## Version Compatibility

- **Python**: 3.11+
- **MCP SDK**: 1.0+
- **Semantic Scholar API**: v1
- **IDEs**: Cursor 0.40+, VS Code 1.85+, Windsurf 1.0+

For updates and changelog, see [GitHub Releases](https://github.com/yourusername/polyhedra/releases).