smry Product

What changed since a cursor — the polling primitive

get_feed_changes

Read-only

One stream, two checkpoint owners. Stateful callers pass cursor:"latest", then store each next_cursor themselves; that first call only establishes the checkpoint and returns no events, so poll again with the returned next_cursor. Stateless scheduled agents pass consumer_id; smry retains the acknowledged position and the consumer's saved where predicate. Pass exactly one. Managed batches repeat until their next_cursor is acknowledged. Every response also carries a capped source-health attention block. To read what is already there rather than what changes next, use query_feeds. Example: {"cursor":"latest","limit":20} Example: {"consumer_id":7,"limit":20}

Input Schema

TableJSON Schema

Name	Required	Description
`limit`	No
`where`	No	Entry selection predicate: fields AND together; omit a field to not filter on it. Source queries/mutations accept only collection and source; use query having for source-health filters.
`cursor`	No	Caller-managed progress: "latest" to start, then the next_cursor you stored.
`fields`	No	Project event entries to these fields; ["id"] is the cheapest diff. id is always included.
`consumer_id`	No	smry-managed progress: poll this consumer's saved predicate. Omit cursor and where.
`attention_limit`	No	Sources listed per attention bucket (default 5; the *_count fields always carry the full tally). Pass up to 500 for the exhaustive roster, 0 for counts only.

Output Schema

TableJSON Schema

Name	Required	Description
`events`	Yes
`object`	Yes
`has_more`	Yes
`attention`	Yes	Source health, capped to five per bucket (stalled: longest silent first; erroring and never_produced: alphabetical); the *_count fields carry the full tally. Query select:"sources" with having for the complete list.
`checkpoint`	Yes
`next_cursor`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed5 schema fields changedSep 13, 2026

addedOutput schema / properties / events / items / additionalProperties
```
Added value: +{}
```
addedOutput schema / properties / events / items / description
```
Added value: +"A single change event."
```
removedOutput schema / properties / events / items / patternProperties
```
Removed value: -{
-  "^(.*)$": {}
-}
```

addedOutput schema / properties / events / items / properties

Added value: +{
+  "entry": {
+    "description": "The entry, projected to the requested `fields`. `id` is always present. entry.classified events also carry the filter's decision, assessment, reason, and filter_revision.",
+    "patternProperties": {
+      "^(.*)$": {}
+    },
+    "type": "object"
+  },
+  "type": {
+    "anyOf": [
+      {
+        "const": "entry.discovered",
+        "type": "string"
+      },
+      {
+        "const": "entry.classified",
+        "type": "string"
+      }
+    ],
+    "description": "entry.discovered: smry saw this entry for the first time. entry.classified: a Collection filter reached a decision about an entry you already have."
+  }
+}

addedOutput schema / properties / events / items / required
```
Added value: +[
+  "type",
+  "entry"
+]
```

Changed12 schema fields changedSep 7, 2026

addedInput schema / properties / attention_limit

Added value: +{
+  "description": "Sources listed per attention bucket (default 5; the *_count fields always carry the full tally). Pass up to 500 for the exhaustive roster, 0 for counts only.",
+  "maximum": 500,
+  "minimum": 0,
+  "type": "integer"
+}

changedInput schema / properties / where / properties / discovered_after / description

Previous value: -"Relative window (\"7d\", \"36h\", \"45m\", \"2w\") or ISO date."New value: +"When smry first saw the entry. Backfill makes old posts look new here, so use it for diffs, not for \"what's new\". Relative window (\"7d\", \"36h\", \"45m\", \"2w\"), \"now\", or ISO date."

changedInput schema / properties / where / properties / discovered_before / description

Previous value: -"Relative window (\"7d\", \"36h\", \"45m\", \"2w\") or ISO date."New value: +"Relative window (\"7d\", \"36h\", \"45m\", \"2w\"), \"now\", or ISO date."

changedInput schema / properties / where / properties / published_after / description

Previous value: -"Relative window (\"7d\", \"36h\", \"45m\", \"2w\") or ISO date."New value: +"When the source published the entry (entries with no publish date are excluded; use top-level since to include them). Relative window (\"7d\", \"36h\", \"45m\", \"2w\"), \"now\", or ISO date."

changedInput schema / properties / where / properties / published_before / description

Previous value: -"Relative window (\"7d\", \"36h\", \"45m\", \"2w\") or ISO date."New value: +"Relative window (\"7d\", \"36h\", \"45m\", \"2w\"), \"now\", or ISO date."

changedInput schema / properties / where / properties / source / properties / exclude / items / description

Previous value: -"A source id (from any v2 response) or its exact title."New value: +"A source id (from any v2 response) or its title. Reads also accept a fragment that matches exactly one title."

changedInput schema / properties / where / properties / source / properties / include / items / description

Previous value: -"A source id (from any v2 response) or its exact title."New value: +"A source id (from any v2 response) or its title. Reads also accept a fragment that matches exactly one title."

addedOutput schema / properties / attention / description

Added value: +"Source health, capped to five per bucket (stalled: longest silent first; erroring and never_produced: alphabetical); the *_count fields carry the full tally. Query select:\"sources\" with having for the complete list."

addedOutput schema / properties / attention / properties / erroring_count

Added value: +{
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}

addedOutput schema / properties / attention / properties / never_produced_count

Added value: +{
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}

addedOutput schema / properties / attention / properties / stalled_count

Added value: +{
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}

changedOutput schema / properties / attention / required

Previous value: -[
-  "stalled",
-  "erroring",
-  "never_produced"
-]New value: +[
+  "stalled",
+  "stalled_count",
+  "erroring",
+  "erroring_count",
+  "never_produced",
+  "never_produced_count"
+]

Changed17 schema fields changedSep 7, 2026

addedInput schema / additionalProperties
```
Added value: +false
```

removedInput schema / properties / consumer_id / anyOf

Removed value: -[
-  {
-    "default": 0,
-    "type": "string"
-  },
-  {
-    "description": "smry-managed progress: poll this consumer's saved predicate. Omit cursor and where.",
-    "maximum": 9007199254740991,
-    "minimum": 1,
-    "type": "integer"
-  }
-]

addedInput schema / properties / consumer_id / maximum
```
Added value: +9007199254740991
```
addedInput schema / properties / consumer_id / minimum
```
Added value: +1
```
addedInput schema / properties / consumer_id / type
```
Added value: +"integer"
```

removedInput schema / properties / limit / anyOf

Removed value: -[
-  {
-    "default": 0,
-    "type": "string"
-  },
-  {
-    "maximum": 500,
-    "minimum": 1,
-    "type": "integer"
-  }
-]

addedInput schema / properties / limit / maximum
```
Added value: +500
```
addedInput schema / properties / limit / minimum
```
Added value: +1
```
addedInput schema / properties / limit / type
```
Added value: +"integer"
```

changedInput schema / properties / where / description

Previous value: -"Selection predicate shared by every feeds operation. All fields AND together; omit a field to not filter on it."New value: +"Entry selection predicate: fields AND together; omit a field to not filter on it. Source queries/mutations accept only collection and source; use query having for source-health filters."

changedInput schema / properties / where / properties / collection / items / anyOf

Previous value: -[
-  {
-    "anyOf": [
-      {
-        "default": 0,
-        "type": "string"
-      },
-      {
-        "maximum": 9007199254740991,
-        "minimum": 1,
-        "type": "integer"
-      }
-    ]
-  },
-  {
-    "maxLength": 100,
-    "minLength": 1,
-    "type": "string"
-  }
-]New value: +[
+  {
+    "maximum": 9007199254740991,
+    "minimum": 1,
+    "type": "integer"
+  },
+  {
+    "maxLength": 100,
+    "minLength": 1,
+    "type": "string"
+  }
+]

addedInput schema / properties / where / properties / collection / minItems
```
Added value: +1
```

changedInput schema / properties / where / properties / source / properties / exclude / items / anyOf

Previous value: -[
-  {
-    "anyOf": [
-      {
-        "default": 0,
-        "type": "string"
-      },
-      {
-        "maximum": 9007199254740991,
-        "minimum": 1,
-        "type": "integer"
-      }
-    ]
-  },
-  {
-    "maxLength": 500,
-    "minLength": 1,
-    "type": "string"
-  }
-]New value: +[
+  {
+    "maximum": 9007199254740991,
+    "minimum": 1,
+    "type": "integer"
+  },
+  {
+    "maxLength": 500,
+    "minLength": 1,
+    "type": "string"
+  }
+]

addedInput schema / properties / where / properties / source / properties / exclude / minItems
```
Added value: +1
```

changedInput schema / properties / where / properties / source / properties / include / items / anyOf

Previous value: -[
-  {
-    "anyOf": [
-      {
-        "default": 0,
-        "type": "string"
-      },
-      {
-        "maximum": 9007199254740991,
-        "minimum": 1,
-        "type": "integer"
-      }
-    ]
-  },
-  {
-    "maxLength": 500,
-    "minLength": 1,
-    "type": "string"
-  }
-]New value: +[
+  {
+    "maximum": 9007199254740991,
+    "minimum": 1,
+    "type": "integer"
+  },
+  {
+    "maxLength": 500,
+    "minLength": 1,
+    "type": "string"
+  }
+]

addedInput schema / properties / where / properties / source / properties / include / minItems
```
Added value: +1
```

changedOutput schema / properties / checkpoint / properties / consumer_id / anyOf

Previous value: -[
-  {
-    "anyOf": [
-      {
-        "default": 0,
-        "type": "string"
-      },
-      {
-        "maximum": 9007199254740991,
-        "minimum": -9007199254740991,
-        "type": "integer"
-      }
-    ]
-  },
-  {
-    "type": "null"
-  }
-]New value: +[
+  {
+    "maximum": 9007199254740991,
+    "minimum": -9007199254740991,
+    "type": "integer"
+  },
+  {
+    "type": "null"
+  }
+]

AddedSep 7, 2026

TDQS

A4.4/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With readOnlyHint/not-destructive already given by annotations, the description still adds critical non-obvious behavior: the first cursor:"latest" call establishes the checkpoint and returns no events, managed batches repeat until acknowledged, and every response includes a capped source-health attention block. These are behavioral traits the annotations cannot convey.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The content is dense but front-loaded, with usage modes stated before examples. The opening metaphor and terms like "smry" and "managed batches" assume prior context, mildly hurting first-read clarity. Nothing is superfluous, so structure is good without being flawless.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return values need not be explained, and the description still flags the attention block. It covers both polling modes and the acknowledge-until-done loop, leaving mostly the nested where/classification semantics to the schema. Complete enough for correct invocation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is high (83%), so the baseline is 3. The description exceeds it by clarifying the cursor-vs-consumer_id dichotomy, noting that a consumer's where predicate is retained server-side, and showing two concrete call examples. It adds real semantic value beyond the schema, though some nested where/field semantics remain schema-only.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a specific operation — returning feed changes since a checkpoint — and explicitly distinguishes itself from query_feeds ("read what is already there rather than what changes next"). The opening metaphor "One stream, two checkpoint owners" is somewhat cryptic, but the body quickly grounds it in concrete modes. Siblings are addressed, so it earns above a 4-level baseline.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It explicitly routes two caller types: stateful callers pass cursor:"latest" and store next_cursor themselves, while stateless scheduled agents pass consumer_id. It states the exclusivity rule ("Pass exactly one") and names query_feeds as the alternative for reading existing content rather than changes. This is close to exhaustive when/when-not guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All smry Product tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?