weavatrix 1.17.4 → 1.17.6

package/skill/references/tools/semantic_link.md

@@ -1,17 +1,20 @@
 # `semantic_link`
 
-Build inferred semantic graph evidence from supplied vectors.
+Rank symbols for a phrase, or link caller-supplied vectors.
 
 ## When to use
 
-Build inferred semantic graph edges from vectors supplied by the caller; Weavatrix does not create embeddings.
+Pass `query` to rank functions, methods, and types by the local deterministic token hash `weavatrix-token-hash-v1` (name, path, and the declaration line). That hash is not a neural embedding, and Weavatrix does not create one.
+
+Pass a non-empty `vectors` array only when you already have embeddings. An empty `vectors` array without `query` is an error and builds no links. When both are set, the supplied vectors are linked and the phrase is not ranked.
 
 ## Inputs
 
-- `vectors` (array<object>, required).
+- `query` (string) — phrase for `weavatrix-token-hash-v1`. Not a neural embedding.
+- `vectors` (array<object>) — caller-supplied embedding rows. Required only when `query` is omitted; an empty array is an error in that case.
 - `min_similarity` (number, min 0, max 100) — 0..1 is a fraction; values above 1 through 100 are percentages.
 - `model` (string).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 - `selection` ("union" | "mutual" | "directed").
 - `top_k` (string).
 
@@ -21,11 +24,10 @@ Build inferred semantic graph edges from vectors supplied by the caller; Weavatr
 {
   "name": "semantic_link",
   "arguments": {
-    "vectors": [{}],
+    "query": "strict manufacturer identity",
     "output_format": "text"
   }
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/shortest_path.md

@@ -11,7 +11,7 @@ Prove the shortest typed dependency path between two exact nodes.
 - `source` (string, required).
 - `target` (string, required).
 - `max_hops` (integer, min 0).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 
 ## Minimal call
 
@@ -26,5 +26,4 @@ Prove the shortest typed dependency path between two exact nodes.
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/trace_api_contract.md

@@ -26,7 +26,7 @@ Keep Kafka, AMQP/RabbitMQ, NATS, SNS/SQS, JMS, GraphQL, and gRPC identities dist
 - `max_source_file_bytes` (integer, min 0).
 - `max_source_files` (integer, min 0).
 - `method` (string).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 - `page_size` (integer, min 0).
 - `path` (string).
 - `per_item_limit` (integer, min 0).
@@ -50,5 +50,4 @@ Keep Kafka, AMQP/RabbitMQ, NATS, SNS/SQS, JMS, GraphQL, and gRPC identities dist
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/trace_endpoint.md

@@ -16,7 +16,7 @@ Resolve one HTTP route and inspect its bounded handler and call neighborhood.
 - `max_excerpts` (integer, min 0).
 - `max_nodes` (integer, min 0).
 - `method` (string).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 
 ## Minimal call
 
@@ -30,5 +30,4 @@ Resolve one HTTP route and inspect its bounded handler and call neighborhood.
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/verified_change.md

@@ -22,7 +22,7 @@ Use the same `task` and base revision in `phase:"plan"` and `phase:"verify"`.
 - `max_data_flow_edges` (integer, min 0).
 - `max_impact_nodes` (integer, min 0).
 - `max_symbols` (integer, min 0).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 - `phase` ("plan" | "verify").
 - `precision` (string).
 - `run_tests` (boolean).
@@ -41,5 +41,4 @@ Use the same `task` and base revision in `phase:"plan"` and `phase:"verify"`.
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/verify_architecture.md

@@ -8,7 +8,7 @@ Check current graph dependencies, cycles, and budgets against the active archite
 
 ## Inputs
 
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 
 ## Minimal call
 
@@ -21,5 +21,4 @@ Check current graph dependencies, cycles, and budgets against the active archite
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/verify_capabilities.md

@@ -11,7 +11,7 @@ Resolve declared capabilities against statically exposed endpoint evidence.
 - `include_classified` (boolean).
 - `include_tests` (boolean).
 - `max_results` (integer, min 0).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 
 ## Minimal call
 
@@ -24,5 +24,4 @@ Resolve declared capabilities against statically exposed endpoint evidence.
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/web3_context.md

@@ -14,7 +14,7 @@ needs result consumers. A live chain is never queried.
 - `label` (string, required) — graph id or unique Web3 member / consumer label.
 - `task` (string, optional) — what the caller intends to change or inspect.
 - `max_related` (number, optional) — related row cap, 1–200, default 24.
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 
 ## Minimal call
 
@@ -29,5 +29,4 @@ needs result consumers. A live chain is never queried.
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/web3_impact.md

@@ -20,7 +20,7 @@ object are not the callee.
 - `provider` (string, optional) — alias for candidate.
 - `task` (string, optional) — selection hint such as `change-event`, `change-input`, or `change-output`.
 - `max_results` (number, optional) — change page size, 1–200, default 50.
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 
 ## Minimal call
 
@@ -34,5 +34,4 @@ object are not the callee.
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/web3_inventory.md

@@ -12,7 +12,7 @@ in the repository. ABI equality is not a deployment proof.
 
 - `path` (string, optional) — repository-relative ABI, artifact, or consumer path fragment.
 - `max_results` (number, optional) — inventory page size, 1–500, default 200.
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 
 ## Minimal call
 
@@ -25,5 +25,4 @@ in the repository. ABI equality is not a deployment proof.
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/web3_trace.md

@@ -16,7 +16,7 @@ the walk finished.
 - `depth` (number, optional) — walk depth, 1–32, default 8.
 - `max_nodes` (number, optional) — page size, 1–200, default 64.
 - `cursor` (string, optional) — opaque page token from a previous page.
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
+- `output_format` ("text" | "json" | "structured", default "json") — text returns one JSON text block of the full payload, not a shorter summary; json returns structured output and mirrors that JSON into text, so the answer is about twice as large; structured drops the mirror and is the smaller payload when the client reads structuredContent.
 
 ## Minimal call
 
@@ -30,5 +30,4 @@ the walk finished.
 }
 ```
 
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.
+Prefer `output_format:"structured"` when the client reads structuredContent. `text` is still the full JSON payload. The live MCP `tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/memory_context.md

@@ -1,29 +0,0 @@
-# `memory_context`
-
-Compile bounded temporal memory context from supplied events.
-
-## When to use
-
-Compile bounded temporal context from events and a request supplied by the caller.
-
-## Inputs
-
-- `events` (array<object>, required).
-- `request` (object, required).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
-
-## Minimal call
-
-```json
-{
-  "name": "memory_context",
-  "arguments": {
-    "events": [{}],
-    "request": {},
-    "output_format": "text"
-  }
-}
-```
-
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/seo_link_suggestions.md

@@ -1,34 +0,0 @@
-# `seo_link_suggestions`
-
-Directional SEO internal-link evidence from supplied page profiles.
-
-## When to use
-
-Generate directional internal-link evidence from caller-supplied page profiles and vectors.
-
-## Inputs
-
-- `pages` (array<object>, required).
-- `vectors` (array<object>, required).
-- `allow_cross_language` (boolean).
-- `min_similarity` (number, min 0, max 100) — 0..1 is a fraction; values above 1 through 100 are percentages.
-- `model` (string).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
-- `selection` ("union" | "mutual" | "directed").
-- `top_k` (string).
-
-## Minimal call
-
-```json
-{
-  "name": "seo_link_suggestions",
-  "arguments": {
-    "pages": [{}],
-    "vectors": [{}],
-    "output_format": "text"
-  }
-}
-```
-
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.

package/skill/references/tools/vector_search.md

@@ -1,31 +0,0 @@
-# `vector_search`
-
-Exact or bounded approximate nearest-neighbor search.
-
-## When to use
-
-Run exact or bounded approximate nearest-neighbor search over vectors supplied by the caller.
-
-## Inputs
-
-- `query` (array<number>, required).
-- `vectors` (array<object>, required).
-- `exact` (boolean).
-- `output_format` ("text" | "json" | "structured", default "json") — text returns the concise text block only; json returns structured output and mirrors it into text for clients that read only content; structured drops that mirror, which is the larger copy, and is safe only where the client reads structuredContent.
-- `top_k` (string).
-
-## Minimal call
-
-```json
-{
-  "name": "vector_search",
-  "arguments": {
-    "query": ["<value>"],
-    "vectors": [{}],
-    "output_format": "text"
-  }
-}
-```
-
-Use `output_format:"text"` for compact agent interaction. The live MCP
-`tools/list` schema remains authoritative for this installed version.