@mastra/mcp-docs-server 1.3.6 → 1.3.7-alpha.1
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/.docs/docs/memory/observational-memory.md +17 -15
- package/.docs/models/index.md +1 -1
- package/.docs/models/providers/edenai.md +4 -3
- package/.docs/models/providers/kilo.md +6 -6
- package/.docs/models/providers/nearai.md +6 -1
- package/.docs/reference/memory/observational-memory.md +3 -3
- package/package.json +4 -4
|
@@ -468,7 +468,7 @@ With default settings, the context window doesn't grow unbounded. It oscillates
|
|
|
468
468
|
3. **Repeat**: History grows from \~6k back toward 30k and shrinks again. Each cycle appends to the observation log, which grows much more slowly than raw history.
|
|
469
469
|
4. **Observations reach 40k**: The Reflector creates a smaller log from the current observations and any earlier reflections.
|
|
470
470
|
|
|
471
|
-
In the normal buffered cycle, raw history oscillates between roughly 6k and 30k tokens. The observation log stays around 40k tokens, however long the conversation runs. These are activation thresholds rather than hard caps, so history can grow past the threshold whenever background buffering doesn't keep pace.
|
|
471
|
+
In the normal buffered cycle, raw history oscillates between roughly 6k and 30k tokens. The observation log stays around 40k tokens, however long the conversation runs. These are activation thresholds rather than hard caps, so history can grow past the threshold whenever background buffering doesn't keep pace. Between the threshold and `blockAfter` (default `1.2`, \~36k tokens), buffering keeps running in the background and the next ready chunk activates without blocking the agent. At `blockAfter`, a synchronous observation runs if no buffered chunk activates. Reflection falls back to a synchronous run above its own `blockAfter` (\~48k tokens).
|
|
472
472
|
|
|
473
473
|
With [`shareTokenBudget`](https://mastra.ai/reference/memory/observational-memory) enabled, the two budgets pool together. While the observation log is small, message history can expand into the unused observation space (up to \~70k tokens with the defaults) before observation triggers. It then shrinks as observations accumulate.
|
|
474
474
|
|
|
@@ -796,23 +796,25 @@ When message tokens reach the `messageTokens` threshold, buffered chunks activat
|
|
|
796
796
|
|
|
797
797
|
Async buffered Observer calls don't generate continuation hints because delayed hints can be stale by activation time. When buffered chunks activate, any previously stored suggested response and current task are cleared. The main agent receives the compressed observations without those hints.
|
|
798
798
|
|
|
799
|
-
When message production outpaces the Observer,
|
|
799
|
+
When message production outpaces the Observer, history can pass `messageTokens` before a buffered chunk is ready. Between `messageTokens` and `blockAfter`, Mastra keeps buffering in the background and activates the chunk on a later step, so the agent doesn't wait on the Observer. A synchronous observation runs if pending tokens still reach `blockAfter` after buffered activation, for example when a large tool result leaves an uncovered tail. Above `blockAfter`, activation may also overshoot the retention target instead of using fewer chunks. Buffered activation usually preserves a minimum remaining context (the smaller of \~1k tokens or the configured retention floor), but a single buffered chunk that covers the whole pending window still activates and can leave less.
|
|
800
|
+
|
|
801
|
+
At `blockAfter`, activation waits a bounded time for a buffered observation that's still being produced so it activates with the rest. If that takes longer, Mastra observes the remaining messages synchronously. A late observation whose messages are all already observed is then dropped instead of stored. Activations triggered by the idle TTL or a model change don't wait.
|
|
800
802
|
|
|
801
803
|
Reflection works similarly: the Reflector runs in the background when observations reach a fraction of the reflection threshold.
|
|
802
804
|
|
|
803
805
|
### Settings
|
|
804
806
|
|
|
805
|
-
| Setting | Default | What it controls
|
|
806
|
-
| ------------------------------------- | ------- |
|
|
807
|
-
| `observation.bufferTokens` | `0.2` | How often to buffer. `0.2` means every 20% of `messageTokens`. With the default 30k threshold, that's roughly every 6k tokens. Can also be an absolute token count (e.g. `5000`).
|
|
808
|
-
| `observation.bufferActivation` | `0.8` | How aggressively to clear the message window on activation. `0.8` means remove enough messages to keep only 20% of `messageTokens` remaining. Lower values keep more message history.
|
|
809
|
-
| `observation.blockAfter` | `1.2` | Safety net if buffering can't keep up. Values from 1 up to (but not including) 100 multiply `messageTokens`. For example, `1.2` creates a threshold of 36k tokens (1.2 × 30k),
|
|
810
|
-
| `activateAfterIdle` | none | Forces buffered observations to activate after a period of inactivity, even before `observation.messageTokens` is reached. Accepts a numeric millisecond value such as `300_000`, duration strings like `"5m"` or `"1hr"`, `"auto"` for a provider-aware prompt cache TTL, or an object of per-provider TTLs such as `{ default: 'auto', anthropic: '1h' }`.
|
|
811
|
-
| `activateOnProviderChange` | `false` | Forces buffered observations to activate when the next step uses a different `provider/model` than the one that produced the latest assistant step. Use this when switching providers or models would invalidate prompt cache reuse.
|
|
812
|
-
| `reflection.bufferActivation` | `0.5` | When to start background reflection. `0.5` means reflection begins when observations reach 50% of the `observationTokens` threshold.
|
|
813
|
-
| `reflection.activateAfterIdle` | none | Opts buffered reflections into idle activation. Reflections don't inherit top-level `activateAfterIdle`.
|
|
814
|
-
| `reflection.activateOnProviderChange` | `false` | Opts buffered reflections into provider-change activation. Reflections don't inherit top-level `activateOnProviderChange`.
|
|
815
|
-
| `reflection.blockAfter` | `1.2` | Safety threshold for reflection. Same value format as observation (absolute values must be greater than `observationTokens`), but above it reflection runs synchronously when no buffered reflection is ready to activate.
|
|
807
|
+
| Setting | Default | What it controls |
|
|
808
|
+
| ------------------------------------- | ------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
|
809
|
+
| `observation.bufferTokens` | `0.2` | How often to buffer. `0.2` means every 20% of `messageTokens`. With the default 30k threshold, that's roughly every 6k tokens. Can also be an absolute token count (e.g. `5000`). |
|
|
810
|
+
| `observation.bufferActivation` | `0.8` | How aggressively to clear the message window on activation. `0.8` means remove enough messages to keep only 20% of `messageTokens` remaining. Lower values keep more message history. |
|
|
811
|
+
| `observation.blockAfter` | `1.2` | Safety net if buffering can't keep up. Values from 1 up to (but not including) 100 multiply `messageTokens`. For example, `1.2` creates a threshold of 36k tokens (1.2 × 30k). Below it, only background buffering and activation run. Observation runs synchronously if pending tokens still reach it after buffered activation. Activation may also overshoot the retention target. Values of 100 or more are absolute token counts (e.g. `50_000`) and must be greater than `messageTokens`. Set `1` to observe synchronously at `messageTokens`. |
|
|
812
|
+
| `activateAfterIdle` | none | Forces buffered observations to activate after a period of inactivity, even before `observation.messageTokens` is reached. Accepts a numeric millisecond value such as `300_000`, duration strings like `"5m"` or `"1hr"`, `"auto"` for a provider-aware prompt cache TTL, or an object of per-provider TTLs such as `{ default: 'auto', anthropic: '1h' }`. |
|
|
813
|
+
| `activateOnProviderChange` | `false` | Forces buffered observations to activate when the next step uses a different `provider/model` than the one that produced the latest assistant step. Use this when switching providers or models would invalidate prompt cache reuse. |
|
|
814
|
+
| `reflection.bufferActivation` | `0.5` | When to start background reflection. `0.5` means reflection begins when observations reach 50% of the `observationTokens` threshold. |
|
|
815
|
+
| `reflection.activateAfterIdle` | none | Opts buffered reflections into idle activation. Reflections don't inherit top-level `activateAfterIdle`. |
|
|
816
|
+
| `reflection.activateOnProviderChange` | `false` | Opts buffered reflections into provider-change activation. Reflections don't inherit top-level `activateOnProviderChange`. |
|
|
817
|
+
| `reflection.blockAfter` | `1.2` | Safety threshold for reflection. Same value format as observation (absolute values must be greater than `observationTokens`), but above it reflection runs synchronously when no buffered reflection is ready to activate. |
|
|
816
818
|
|
|
817
819
|
If you're relying on prompt caching, set `activateAfterIdle` to `"auto"` or to a specific cache TTL. That way, once a thread has been idle long enough for the cache to expire, the next request can activate buffered observations first and send a smaller compressed context window.
|
|
818
820
|
|
|
@@ -909,9 +911,9 @@ const memory = new Memory({
|
|
|
909
911
|
})
|
|
910
912
|
```
|
|
911
913
|
|
|
912
|
-
- `previousObserverTokens:
|
|
914
|
+
- `previousObserverTokens: false` → default. Sends the full previous observations.
|
|
915
|
+
- `previousObserverTokens: 2000` → keeps \~2k tokens of recent observations.
|
|
913
916
|
- `previousObserverTokens: 0` → omit previous observations completely.
|
|
914
|
-
- `previousObserverTokens: false` → disable truncation and keep full previous observations.
|
|
915
917
|
|
|
916
918
|
## Hooks
|
|
917
919
|
|
package/.docs/models/index.md
CHANGED
|
@@ -4,7 +4,7 @@
|
|
|
4
4
|
|
|
5
5
|
# Model Providers
|
|
6
6
|
|
|
7
|
-
Mastra provides a unified interface for working with LLMs across multiple providers, giving you access to
|
|
7
|
+
Mastra provides a unified interface for working with LLMs across multiple providers, giving you access to 7828 models from 214 providers through a single API.
|
|
8
8
|
|
|
9
9
|
## Features
|
|
10
10
|
|
|
@@ -4,7 +4,7 @@
|
|
|
4
4
|
|
|
5
5
|
# Eden AI
|
|
6
6
|
|
|
7
|
-
Access
|
|
7
|
+
Access 303 Eden AI models through Mastra's model router. Authentication is handled automatically using the `EDENAI_API_KEY` environment variable.
|
|
8
8
|
|
|
9
9
|
Learn more in the [Eden AI documentation](https://docs.edenai.co).
|
|
10
10
|
|
|
@@ -152,10 +152,11 @@ for await (const chunk of stream) {
|
|
|
152
152
|
| `edenai/fireworks_ai/accounts/fireworks/models/gpt-oss-120b` | 131K | | | | | | $0.15 | $0.60 |
|
|
153
153
|
| `edenai/fireworks_ai/accounts/fireworks/models/inkling` | 1.0M | | | | | | $1 | $4 |
|
|
154
154
|
| `edenai/fireworks_ai/gpt-oss-120b` | 131K | | | | | | $0.15 | $0.60 |
|
|
155
|
-
| `edenai/flexai/DeepSeek-V4-Flash-0731` | 1.0M | | | | | | $0.
|
|
156
|
-
| `edenai/flexai/DeepSeek-V4.1-Flash` | 1.0M | | | | | | $0.
|
|
155
|
+
| `edenai/flexai/DeepSeek-V4-Flash-0731` | 1.0M | | | | | | $0.02 | $0.27 |
|
|
156
|
+
| `edenai/flexai/DeepSeek-V4.1-Flash` | 1.0M | | | | | | $0.09 | $0.18 |
|
|
157
157
|
| `edenai/flexai/gpt-oss-120b` | 131K | | | | | | $0.03 | $0.17 |
|
|
158
158
|
| `edenai/flexai/gpt-oss-20b` | 131K | | | | | | $0.02 | $0.10 |
|
|
159
|
+
| `edenai/flexai/Muse-Glimmer-30B` | 131K | | | | | | $0.35 | $2 |
|
|
159
160
|
| `edenai/flexai/Step-3.7-Flash` | 262K | | | | | | $0.20 | $1 |
|
|
160
161
|
| `edenai/google/deep-research-max-preview-04-2026` | 1.0M | | | | | | $2 | $12 |
|
|
161
162
|
| `edenai/google/deep-research-preview-04-2026` | 1.0M | | | | | | $2 | $12 |
|
|
@@ -43,11 +43,11 @@ for await (const chunk of stream) {
|
|
|
43
43
|
| `kilo/~anthropic/claude-opus-latest` | 1.0M | | | | | | $4 | $20 |
|
|
44
44
|
| `kilo/~anthropic/claude-sonnet-latest` | 1.0M | | | | | | $2 | $10 |
|
|
45
45
|
| `kilo/~deepseek/deepseek-flash-latest` | 1.0M | | | | | | $0.03 | $0.14 |
|
|
46
|
-
| `kilo/~deepseek/deepseek-pro-latest` | 1.0M | | | | | | $0.
|
|
47
|
-
| `kilo/~deepseek/deepseek-v4-flash-latest` | 1.0M | | | | | | $0.02 | $0.
|
|
46
|
+
| `kilo/~deepseek/deepseek-pro-latest` | 1.0M | | | | | | $0.21 | $0.63 |
|
|
47
|
+
| `kilo/~deepseek/deepseek-v4-flash-latest` | 1.0M | | | | | | $0.02 | $0.90 |
|
|
48
48
|
| `kilo/~google/gemini-flash-latest` | 1.0M | | | | | | $0.75 | $4 |
|
|
49
49
|
| `kilo/~google/gemini-pro-latest` | 1.0M | | | | | | $2 | $12 |
|
|
50
|
-
| `kilo/~moonshotai/kimi-latest` | 1.0M | | | | | | $0.
|
|
50
|
+
| `kilo/~moonshotai/kimi-latest` | 1.0M | | | | | | $0.72 | $17 |
|
|
51
51
|
| `kilo/~openai/gpt-astra-latest` | 1.1M | | | | | | $10 | $50 |
|
|
52
52
|
| `kilo/~openai/gpt-luna-latest` | 1.1M | | | | | | $0.10 | $0.50 |
|
|
53
53
|
| `kilo/~openai/gpt-mini-latest` | 400K | | | | | | $0.75 | $5 |
|
|
@@ -55,7 +55,7 @@ for await (const chunk of stream) {
|
|
|
55
55
|
| `kilo/~openai/gpt-terra-latest` | 1.1M | | | | | | $2 | $12 |
|
|
56
56
|
| `kilo/~x-ai/grok-latest` | 500K | | | | | | $2 | $6 |
|
|
57
57
|
| `kilo/~z-ai/glm-flash-latest` | 1.0M | | | | | | $0.04 | $0.50 |
|
|
58
|
-
| `kilo/~z-ai/glm-latest` | 1.0M | | | | | | $0.
|
|
58
|
+
| `kilo/~z-ai/glm-latest` | 1.0M | | | | | | $0.08 | $5 |
|
|
59
59
|
| `kilo/aion-labs/aion-2.0` | 131K | | | | | | $0.80 | $2 |
|
|
60
60
|
| `kilo/aion-labs/aion-3.0` | 131K | | | | | | $3 | $6 |
|
|
61
61
|
| `kilo/aion-labs/aion-3.0-mini` | 131K | | | | | | $0.70 | $1 |
|
|
@@ -380,9 +380,9 @@ for await (const chunk of stream) {
|
|
|
380
380
|
| `kilo/tencent/hy-mt2-1.8b` | 8K | | | | | | $0.04 | $0.18 |
|
|
381
381
|
| `kilo/tencent/hy-mt2-30b-a3b` | 8K | | | | | | $0.07 | $0.29 |
|
|
382
382
|
| `kilo/tencent/hy-mt2-7b` | 8K | | | | | | $0.07 | $0.29 |
|
|
383
|
-
| `kilo/tencent/hy3` | 262K | | | | | | $0.
|
|
383
|
+
| `kilo/tencent/hy3` | 262K | | | | | | $0.08 | $0.33 |
|
|
384
384
|
| `kilo/tencent/hy3-preview` | 262K | | | | | | $0.18 | $0.60 |
|
|
385
|
-
| `kilo/tencent/hy4-preview` | 1.0M | | | | | | $0.
|
|
385
|
+
| `kilo/tencent/hy4-preview` | 1.0M | | | | | | $0.75 | $2 |
|
|
386
386
|
| `kilo/thedrummer/cydonia-24b-v4.1` | 131K | | | | | | $0.30 | $0.50 |
|
|
387
387
|
| `kilo/thedrummer/skyfall-36b-v2` | 33K | | | | | | $0.55 | $0.80 |
|
|
388
388
|
| `kilo/thedrummer/unslopnemo-12b` | 1.0M | | | | | | $0.40 | $0.40 |
|
|
@@ -4,7 +4,7 @@
|
|
|
4
4
|
|
|
5
5
|
# NEAR AI Cloud
|
|
6
6
|
|
|
7
|
-
Access
|
|
7
|
+
Access 56 NEAR AI Cloud models through Mastra's model router. Authentication is handled automatically using the `NEARAI_API_KEY` environment variable.
|
|
8
8
|
|
|
9
9
|
Learn more in the [NEAR AI Cloud documentation](https://docs.near.ai/).
|
|
10
10
|
|
|
@@ -41,15 +41,19 @@ for await (const chunk of stream) {
|
|
|
41
41
|
| `nearai/anthropic/claude-fable-5` | 1.0M | | | | | | $10 | $50 |
|
|
42
42
|
| `nearai/anthropic/claude-fable-5-1` | 1.0M | | | | | | $10 | $50 |
|
|
43
43
|
| `nearai/anthropic/claude-haiku-4-5` | 200K | | | | | | $1 | $5 |
|
|
44
|
+
| `nearai/anthropic/claude-haiku-5-5` | 1.0M | | | | | | $0.10 | $0.50 |
|
|
44
45
|
| `nearai/anthropic/claude-opus-4-6` | 200K | | | | | | $5 | $25 |
|
|
45
46
|
| `nearai/anthropic/claude-opus-4-7` | 1.0M | | | | | | $5 | $25 |
|
|
46
47
|
| `nearai/anthropic/claude-opus-4-8` | 1.0M | | | | | | $5 | $25 |
|
|
47
48
|
| `nearai/anthropic/claude-opus-5` | 1.0M | | | | | | $5 | $25 |
|
|
49
|
+
| `nearai/anthropic/claude-opus-5-5` | 1.0M | | | | | | $4 | $20 |
|
|
48
50
|
| `nearai/anthropic/claude-sonnet-4-5` | 200K | | | | | | $3 | $15 |
|
|
49
51
|
| `nearai/anthropic/claude-sonnet-4-6` | 1.0M | | | | | | $3 | $15 |
|
|
50
52
|
| `nearai/anthropic/claude-sonnet-5` | 1.0M | | | | | | $2 | $10 |
|
|
53
|
+
| `nearai/anthropic/claude-sonnet-5-5` | 1.0M | | | | | | $2 | $10 |
|
|
51
54
|
| `nearai/black-forest-labs/FLUX.2-klein-4B` | 128K | | | | | | $1 | $1 |
|
|
52
55
|
| `nearai/deepseek/deepseek-v3.2` | 128K | | | | | | $1 | $1 |
|
|
56
|
+
| `nearai/deepseek/deepseek-v4.1-flash` | 1.0M | | | | | | $0.30 | $1 |
|
|
53
57
|
| `nearai/google/gemini-2.5-flash` | 1.0M | | | | | | $0.30 | $3 |
|
|
54
58
|
| `nearai/google/gemini-2.5-flash-lite` | 1.0M | | | | | | $0.10 | $0.40 |
|
|
55
59
|
| `nearai/google/gemini-2.5-pro` | 1.0M | | | | | | $1 | $10 |
|
|
@@ -88,6 +92,7 @@ for await (const chunk of stream) {
|
|
|
88
92
|
| `nearai/qwen/qwen3.7-max` | 1.0M | | | | | | $3 | $8 |
|
|
89
93
|
| `nearai/Qwen/Qwen3.8-27B` | 262K | | | | | | $0.04 | $1 |
|
|
90
94
|
| `nearai/x-ai/grok-4.6` | 500K | | | | | | $2 | $6 |
|
|
95
|
+
| `nearai/x-ai/grok-4.7` | 500K | | | | | | $2 | $6 |
|
|
91
96
|
| `nearai/z-ai/glm-5.3-flash` | 1.0M | | | | | | $0.15 | $0.50 |
|
|
92
97
|
|
|
93
98
|
Model availability, capabilities, context windows, and pricing are sourced from [models.dev](https://models.dev) and may change.
|
|
@@ -99,7 +99,7 @@ OM performs thresholding with fast local token estimation. Text uses `tokenx`, a
|
|
|
99
99
|
|
|
100
100
|
**observation.activateOnProviderChange** (`boolean`): Force buffered observations to activate when the actor provider or model changes. If unset, the top-level activateOnProviderChange value is used for observations. Currently only applied when using the standalone ObservationalMemory class; new Memory(...) applies the top-level activateOnProviderChange only.
|
|
101
101
|
|
|
102
|
-
**observation.blockAfter** (`number`): Safety net for when background buffering can't keep up. Values from 1 up to (but not including) 100 are multipliers of messageTokens: 1.2 resolves to 120% of the threshold (36k tokens with the default 30k). Values of 100 or more are absolute token counts and must be greater than messageTokens. Above this point, activation uses the smallest set of buffered chunks that reaches the retention target, even when that overshoots the target by more than the usual safeguard allows. It never activates more chunks than are needed to reach the retention target, so it removes only slightly more history than a normal activation. It changes the result only when the retention floor is above roughly 20,000 tokens; with the default bufferActivation (a 6k floor) it has no observable effect. Activation usually keeps a minimum remaining context (the smaller of 1000 tokens or the retention floor), but a single buffered chunk that covers the whole pending window still activates and can leave less.
|
|
102
|
+
**observation.blockAfter** (`number`): Safety net for when background buffering can't keep up. Values from 1 up to (but not including) 100 are multipliers of messageTokens: 1.2 resolves to 120% of the threshold (36k tokens with the default 30k). Multipliers apply to a per-thread messageTokens override when one is set. Values of 100 or more are absolute token counts and must be greater than messageTokens. Between messageTokens and blockAfter, only async buffering and activation run: reaching the threshold without a buffered chunk to activate starts background buffering, and the chunk activates on a later step. A synchronous (blocking) observation runs if pending tokens still reach blockAfter after buffered activation. All remaining buffered chunks are activated first, so the synchronous observation only covers messages no chunk has observed. Set blockAfter: 1 to observe synchronously at messageTokens. Above this point, activation also uses the smallest set of buffered chunks that reaches the retention target, even when that overshoots the target by more than the usual safeguard allows. It never activates more chunks than are needed to reach the retention target, so it removes only slightly more history than a normal activation. It changes the result only when the retention floor is above roughly 20,000 tokens; with the default bufferActivation (a 6k floor) it has no observable effect. Activation usually keeps a minimum remaining context (the smaller of 1000 tokens or the retention floor), but a single buffered chunk that covers the whole pending window still activates and can leave less. Only relevant when bufferTokens is set. Defaults to 1.2 when async buffering is enabled.
|
|
103
103
|
|
|
104
104
|
**observation.previousObserverTokens** (`number | false`): Optional token budget for the observer's previous-observations context. When set to a number, the observations passed to the Observer agent are tail-truncated to fit within this budget while keeping the newest observations and preserving highlighted 🔴 items when possible. When a buffered reflection is pending, the already-reflected observation lines are automatically replaced with the reflection summary before truncation. Set to 0 to omit previous observations entirely, or false to disable truncation explicitly.
|
|
105
105
|
|
|
@@ -371,7 +371,7 @@ export const agent = new Agent({
|
|
|
371
371
|
|
|
372
372
|
Async buffering is **enabled by default**. It pre-computes observations in the background as the conversation grows: when the `messageTokens` threshold is reached, buffered observations activate instantly with no blocking LLM call.
|
|
373
373
|
|
|
374
|
-
The lifecycle follows **buffer → activate → remove messages → repeat**. Background Observer calls run at `bufferTokens` intervals and produce chunks of observations. At the threshold, activation moves observations into the log and removes raw messages from context.
|
|
374
|
+
The lifecycle follows **buffer → activate → remove messages → repeat**. Background Observer calls run at `bufferTokens` intervals and produce chunks of observations. At the threshold, activation moves observations into the log and removes raw messages from context. Between the threshold and `blockAfter`, buffering continues in the background and the next ready chunk activates on a later step. A synchronous observation runs if pending tokens still reach `blockAfter` after all buffered chunks have been activated. Above `blockAfter`, activation may also overshoot the retention target instead of activating fewer chunks.
|
|
375
375
|
|
|
376
376
|
Default settings:
|
|
377
377
|
|
|
@@ -402,7 +402,7 @@ export const agent = new Agent({
|
|
|
402
402
|
bufferTokens: 5_000,
|
|
403
403
|
// Activate to retain 30% of threshold
|
|
404
404
|
bufferActivation: 0.7,
|
|
405
|
-
//
|
|
405
|
+
// Observe synchronously at 1.5x the threshold if no buffered chunk is ready
|
|
406
406
|
blockAfter: 1.5,
|
|
407
407
|
},
|
|
408
408
|
reflection: {
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@mastra/mcp-docs-server",
|
|
3
|
-
"version": "1.3.
|
|
3
|
+
"version": "1.3.7-alpha.1",
|
|
4
4
|
"description": "MCP server for accessing Mastra.ai documentation, changelogs, and news.",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"main": "dist/index.js",
|
|
@@ -27,7 +27,7 @@
|
|
|
27
27
|
"local-pkg": "^1.1.2",
|
|
28
28
|
"zod": "^4.6.4",
|
|
29
29
|
"@mastra/mcp": "^2.3.0",
|
|
30
|
-
"@mastra/core": "1.
|
|
30
|
+
"@mastra/core": "1.77.0-alpha.0"
|
|
31
31
|
},
|
|
32
32
|
"devDependencies": {
|
|
33
33
|
"@hono/node-server": "^2.0.0",
|
|
@@ -43,9 +43,9 @@
|
|
|
43
43
|
"tsx": "^4.23.1",
|
|
44
44
|
"typescript": "^7.0.2",
|
|
45
45
|
"vitest": "4.1.11",
|
|
46
|
-
"@internal/lint": "0.0.142",
|
|
47
46
|
"@internal/types-builder": "0.0.117",
|
|
48
|
-
"@
|
|
47
|
+
"@internal/lint": "0.0.142",
|
|
48
|
+
"@mastra/core": "1.77.0-alpha.0"
|
|
49
49
|
},
|
|
50
50
|
"homepage": "https://mastra.ai",
|
|
51
51
|
"repository": {
|