search-strategy

Name: search-strategy
Author: anthropics

4.3kinstalls2.7kforks

How to install search-strategy?

anthropics/knowledge-work-plugins/search-strategy

$npx -y skills add anthropics/knowledge-work-plugins --skill search-strategy

Installs into the current project.

Summary

Query decomposition and multi-source search orchestration. Breaks natural language questions into targeted searches per source, translates queries into source-specific syntax, ranks results by relevance, and handles ambiguity and fallback strategies.

Files · 1

View on GitHub

SKILL.md

1	# Search Strategy
2
3	> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../../CONNECTORS.md).
4
5	The core intelligence behind enterprise search. Transforms a single natural language question into parallel, source-specific searches and produces ranked, deduplicated results.
6
7	## The Goal
8
9	Turn this:
10	```
11	"What did we decide about the API migration timeline?"
12	```
13
14	Into targeted searches across every connected source:
15	```
16	~~chat: "API migration timeline decision" (semantic) + "API migration" in:#engineering after:2025-01-01
17	~~knowledge base: semantic search "API migration timeline decision"
18	~~project tracker: text search "API migration" in relevant workspace
19	```
20
21	Then synthesize the results into a single coherent answer.
22
23	## Query Decomposition
24
25	### Step 1: Identify Query Type
26
27	Classify the user's question to determine search strategy:
28
29	\| Query Type \| Example \| Strategy \|
30	\|-----------\|---------\|----------\|
31	\| Decision \| "What did we decide about X?" \| Prioritize conversations (~~chat, email), look for conclusion signals \|
32	\| Status \| "What's the status of Project Y?" \| Prioritize recent activity, task trackers, status updates \|
33	\| Document \| "Where's the spec for Z?" \| Prioritize Drive, wiki, shared docs \|
34	\| Person \| "Who's working on X?" \| Search task assignments, message authors, doc collaborators \|
35	\| Factual \| "What's our policy on X?" \| Prioritize wiki, official docs, then confirmatory conversations \|
36	\| Temporal \| "When did X happen?" \| Search with broad date range, look for timestamps \|
37	\| Exploratory \| "What do we know about X?" \| Broad search across all sources, synthesize \|
38
39	### Step 2: Extract Search Components
40
41	From the query, extract:
42
43	- Keywords: Core terms that must appear in results
44	- Entities: People, projects, teams, tools (use memory system if available)
45	- Intent signals: Decision words, status words, temporal markers
46	- Constraints: Time ranges, source hints, author filters
47	- Negations: Things to exclude
48
49	### Step 3: Generate Sub-Queries Per Source
50
51	For each available source, create one or more targeted queries:
52
53	Prefer semantic search for:
54	- Conceptual questions ("What do we think about...")
55	- Questions where exact keywords are unknown
56	- Exploratory queries
57
58	Prefer keyword search for:
59	- Known terms, project names, acronyms
60	- Exact phrases the user quoted
61	- Filter-heavy queries (from:, in:, after:)
62
63	Generate multiple query variants when the topic might be referred to differently:
64	```
65	User: "Kubernetes setup"
66	Queries: "Kubernetes", "k8s", "cluster", "container orchestration"
67	```
68
69	## Source-Specific Query Translation
70
71	### ~~chat
72
73	Semantic search (natural language questions):
74	```
75	query: "What is the status of project aurora?"
76	```
77
78	Keyword search:
79	```
80	query: "project aurora status update"
81	query: "aurora in:#engineering after:2025-01-15"
82	query: "from:<@UserID> aurora"
83	```
84
85	Filter mapping:
86	\| Enterprise filter \| ~~chat syntax \|
87	\|------------------\|--------------\|
88	\| `from:sarah` \| `from:sarah` or `from:<@USERID>` \|
89	\| `in:engineering` \| `in:engineering` \|
90	\| `after:2025-01-01` \| `after:2025-01-01` \|
91	\| `before:2025-02-01` \| `before:2025-02-01` \|
92	\| `type:thread` \| `is:thread` \|
93	\| `type:file` \| `has:file` \|
94
95	### ~~knowledge base (Wiki)
96
97	Semantic search — Use for conceptual queries:
98	```
99	descriptive_query: "API migration timeline and decision rationale"
100	```
101
102	Keyword search — Use for exact terms:
103	```
104	query: "API migration"
105	query: "\"API migration timeline\"" (exact phrase)
106	```
107
108	### ~~project tracker
109
110	Task search:
111	```
112	text: "API migration"
113	workspace: [workspace_id]
114	completed: false (for status queries)
115	assignee_any: "me" (for "my tasks" queries)
116	```
117
118	Filter mapping:
119	\| Enterprise filter \| ~~project tracker parameter \|
120	\|------------------\|----------------\|
121	\| `from:sarah` \| `assignee_any` or `created_by_any` \|
122	\| `after:2025-01-01` \| `modified_on_after: "2025-01-01"` \|
123	\| `type:milestone` \| `resource_subtype: "milestone"` \|
124
125	## Result Ranking
126
127	### Relevance Scoring
128
129	Score each result on these factors (weighted by query type):
130
131	\| Factor \| Weight (Decision) \| Weight (Status) \| Weight (Document) \| Weight (Factual) \|
132	\|--------\|-------------------\|------------------\|--------------------\|-------------------\|
133	\| Keyword match \| 0.3 \| 0.2 \| 0.4 \| 0.3 \|
134	\| Freshness \| 0.3 \| 0.4 \| 0.2 \| 0.1 \|
135	\| Authority \| 0.2 \| 0.1 \| 0.3 \| 0.4 \|
136	\| Completeness \| 0.2 \| 0.3 \| 0.1 \| 0.2 \|
137
138	### Authority Hierarchy
139
140	Depends on query type:
141
142	For factual/policy questions:
143	```
144	Wiki/Official docs > Shared documents > Email announcements > Chat messages
145	```
146
147	For "what happened" / decision questions:
148	```

People who liked this also viewed

6 picks

Type

search-strategy

4.3kinstalls2.7kforks

View on GitHub

How to install search-strategy?

anthropics/knowledge-work-plugins/search-strategy

$npx -y skills add anthropics/knowledge-work-plugins --skill search-strategy

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Search Strategy
2
3	> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../../CONNECTORS.md).
4
5	The core intelligence behind enterprise search. Transforms a single natural language question into parallel, source-specific searches and produces ranked, deduplicated results.
6
7	## The Goal
8
9	Turn this:
10	```
11	"What did we decide about the API migration timeline?"
12	```
13
14	Into targeted searches across every connected source:
15	```
16	~~chat: "API migration timeline decision" (semantic) + "API migration" in:#engineering after:2025-01-01
17	~~knowledge base: semantic search "API migration timeline decision"
18	~~project tracker: text search "API migration" in relevant workspace
19	```
20
21	Then synthesize the results into a single coherent answer.
22
23	## Query Decomposition
24
25	### Step 1: Identify Query Type
26
27	Classify the user's question to determine search strategy:
28
29	\| Query Type \| Example \| Strategy \|
30	\|-----------\|---------\|----------\|
31	\| Decision \| "What did we decide about X?" \| Prioritize conversations (~~chat, email), look for conclusion signals \|
32	\| Status \| "What's the status of Project Y?" \| Prioritize recent activity, task trackers, status updates \|
33	\| Document \| "Where's the spec for Z?" \| Prioritize Drive, wiki, shared docs \|
34	\| Person \| "Who's working on X?" \| Search task assignments, message authors, doc collaborators \|
35	\| Factual \| "What's our policy on X?" \| Prioritize wiki, official docs, then confirmatory conversations \|
36	\| Temporal \| "When did X happen?" \| Search with broad date range, look for timestamps \|
37	\| Exploratory \| "What do we know about X?" \| Broad search across all sources, synthesize \|
38
39	### Step 2: Extract Search Components
40
41	From the query, extract:
42
43	- Keywords: Core terms that must appear in results
44	- Entities: People, projects, teams, tools (use memory system if available)
45	- Intent signals: Decision words, status words, temporal markers
46	- Constraints: Time ranges, source hints, author filters
47	- Negations: Things to exclude
48
49	### Step 3: Generate Sub-Queries Per Source
50
51	For each available source, create one or more targeted queries:
52
53	Prefer semantic search for:
54	- Conceptual questions ("What do we think about...")
55	- Questions where exact keywords are unknown
56	- Exploratory queries
57
58	Prefer keyword search for:
59	- Known terms, project names, acronyms
60	- Exact phrases the user quoted
61	- Filter-heavy queries (from:, in:, after:)
62
63	Generate multiple query variants when the topic might be referred to differently:
64	```
65	User: "Kubernetes setup"
66	Queries: "Kubernetes", "k8s", "cluster", "container orchestration"
67	```
68
69	## Source-Specific Query Translation
70
71	### ~~chat
72
73	Semantic search (natural language questions):
74	```
75	query: "What is the status of project aurora?"
76	```
77
78	Keyword search:
79	```
80	query: "project aurora status update"
81	query: "aurora in:#engineering after:2025-01-15"
82	query: "from:<@UserID> aurora"
83	```
84
85	Filter mapping:
86	\| Enterprise filter \| ~~chat syntax \|
87	\|------------------\|--------------\|
88	\| `from:sarah` \| `from:sarah` or `from:<@USERID>` \|
89	\| `in:engineering` \| `in:engineering` \|
90	\| `after:2025-01-01` \| `after:2025-01-01` \|
91	\| `before:2025-02-01` \| `before:2025-02-01` \|
92	\| `type:thread` \| `is:thread` \|
93	\| `type:file` \| `has:file` \|
94
95	### ~~knowledge base (Wiki)
96
97	Semantic search — Use for conceptual queries:
98	```
99	descriptive_query: "API migration timeline and decision rationale"
100	```
101
102	Keyword search — Use for exact terms:
103	```
104	query: "API migration"
105	query: "\"API migration timeline\"" (exact phrase)
106	```
107
108	### ~~project tracker
109
110	Task search:
111	```
112	text: "API migration"
113	workspace: [workspace_id]
114	completed: false (for status queries)
115	assignee_any: "me" (for "my tasks" queries)
116	```
117
118	Filter mapping:
119	\| Enterprise filter \| ~~project tracker parameter \|
120	\|------------------\|----------------\|
121	\| `from:sarah` \| `assignee_any` or `created_by_any` \|
122	\| `after:2025-01-01` \| `modified_on_after: "2025-01-01"` \|
123	\| `type:milestone` \| `resource_subtype: "milestone"` \|
124
125	## Result Ranking
126
127	### Relevance Scoring
128
129	Score each result on these factors (weighted by query type):
130
131	\| Factor \| Weight (Decision) \| Weight (Status) \| Weight (Document) \| Weight (Factual) \|
132	\|--------\|-------------------\|------------------\|--------------------\|-------------------\|
133	\| Keyword match \| 0.3 \| 0.2 \| 0.4 \| 0.3 \|
134	\| Freshness \| 0.3 \| 0.4 \| 0.2 \| 0.1 \|
135	\| Authority \| 0.2 \| 0.1 \| 0.3 \| 0.4 \|
136	\| Completeness \| 0.2 \| 0.3 \| 0.1 \| 0.2 \|
137
138	### Authority Hierarchy
139
140	Depends on query type:
141
142	For factual/policy questions:
143	```
144	Wiki/Official docs > Shared documents > Email announcements > Chat messages
145	```
146
147	For "what happened" / decision questions:
148	```

People who liked this also viewed

6 picks

Type