Skip to main content
Glama
Sign In

Jam

by dev.jam
OverviewDiscussionsAdmin

getNetworkRequests

Read-onlyIdempotent

The Jam's network requests as text, one per line: [M:SS #index] METHOD url → status (initiator, type, duration), with the error, bodies and headers under it. index reads just the requests timeline lines name, with their bodies. Other filters are arrays, OR within one and AND across: statusCode (['4xx','5xx'] for failures), method (['POST','PUT','PATCH','DELETE'] for writes), host, contentType. Only failed requests carry their response body unless bodies:'all' (every body, cut at 4KB: filter first). headers:true adds request and response headers, secret values redacted at capture. The first line gives the match count and the UTC time of [0:00], to match backend logs.

Input Schema

NameRequiredDescriptionDefault
hostNoSubstrings, e.g. ['api.example.com']
toMsNoOnly events at or before this ms into the recording
afterNoFrom the previous page's `More:` line
indexNoEvent indexes from the timeline's `[M:SS #index]`; bodies included
jamIdYesJam identifier — a UUID or a jam.dev URL
limitNoMax events, default 200
bodiesNo'errors' (default): failed requests' bodies; 'all': every body
formatNo'json': the rows with every stored column, for digging into one event or window
fromMsNoOnly events at or after this ms into the recording
intentNoIn 15-30 English words, your larger task and why this call helps it. People by role ("a customer"); no argument values or secrets.
methodNoe.g. ['POST','PUT']
headersNoAdd headers, about 600 bytes a request
llm_modelNoYour exact model id from your system prompt, or "unknown". Never guess.
statusCodeNoe.g. ['500'] or ['4xx','5xx']
contentTypeNoSubstrings, e.g. ['json']

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed16 schema fields changed
    • changedInput schema / properties / after / description
      Previous value: -"The previous page's nextCursor"New value: +"From the previous page's `More:` line"
    • changedInput schema / properties / bodies / description
      Previous value: -"Which response bodies to include. 'errors' (default) returns bodies only for failed requests; 'all' returns every body. Both are truncated to 4KB per body."New value: +"'errors' (default): failed requests' bodies; 'all': every body"
    • changedInput schema / properties / contentType / description
      Previous value: -"Filter by content-type substrings, e.g. ['application/json']"New value: +"Substrings, e.g. ['json']"
    • addedInput schema / properties / format
      Added value: +{
+  "description": "'json': the rows with every stored column, for digging into one event or window",
+  "enum": [
+    "text",
+    "json"
+  ],
+  "type": "string"
+}
    • addedInput schema / properties / fromMs
      Added value: +{
+  "description": "Only events at or after this ms into the recording",
+  "minimum": 0,
+  "type": "integer"
+}
    • changedInput schema / properties / headers / description
      Previous value: -"Include request and response headers (default false). Auth and cookie headers are included; the Jam extension redacts their secret values at capture. Headers average about 600 bytes per request, so filter first."New value: +"Add headers, about 600 bytes a request"
    • changedInput schema / properties / host / description
      Previous value: -"Filter by host substrings, e.g. ['api.example.com']"New value: +"Substrings, e.g. ['api.example.com']"
    • addedInput schema / properties / index
      Added value: +{
+  "description": "Event indexes from the timeline's `[M:SS #index]`; bodies included",
+  "items": {
+    "minimum": 0,
+    "type": "integer"
+  },
+  "maxItems": 500,
+  "minItems": 1,
+  "type": "array"
+}
    • changedInput schema / properties / intent / description
      Previous value: -"In 15-30 words, say what larger task you are working on and why this tool helps it. Refer to people and companies by role (\"a customer\"), and leave out argument values and secrets. Use English. Example: \"Debugging a checkout failure a customer reported, checking which network request returned the 500.\""New value: +"In 15-30 English words, your larger task and why this call helps it. People by role (\"a customer\"); no argument values or secrets."
    • changedInput schema / properties / limit / description
      Previous value: -"Maximum number of events to return"New value: +"Max events, default 200"
    • addedInput schema / properties / limit / minimum
      Added value: +1
    • changedInput schema / properties / limit / type
      Previous value: -"number"New value: +"integer"
    • changedInput schema / properties / llm_model / description
      Previous value: -"The exact model identifier you are running as, from your system prompt or environment (e.g. \"claude-opus-4-8\", \"gpt-5.2\"). Pass \"unknown\" if you are not certain; never guess."New value: +"Your exact model id from your system prompt, or \"unknown\". Never guess."
    • changedInput schema / properties / method / description
      Previous value: -"Filter by HTTP methods, e.g. ['POST','PUT','PATCH','DELETE'] for mutations"New value: +"e.g. ['POST','PUT']"
    • changedInput schema / properties / statusCode / description
      Previous value: -"Filter by status codes and/or classes, e.g. ['500'] or ['4xx','5xx'] for all failures"New value: +"e.g. ['500'] or ['4xx','5xx']"
    • addedInput schema / properties / toMs
      Added value: +{
+  "description": "Only events at or before this ms into the recording",
+  "minimum": 0,
+  "type": "integer"
+}
  2. First observed

TDQS

A3.9/5.0
Behavior4/5

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

Annotations already certify readOnly/idempotent/non-destructive, yet the description still adds real behavioral detail: redaction of secret header values at capture, a 4KB cap on bodies, the default of bodies only for failed requests, and the `More:` pagination line referenced by the `after` param. Only response size/rate expectations are absent.

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

Conciseness3/5

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

It is dense and front-loads the output format, which is right for a 15-param tool, but several clauses are garbled or run together, e.g. "`index` reads just the requests timeline lines name, with their bodies" and "with the error, bodies and headers under it", forcing the reader to re-parse rather than shortening.

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?

No output schema exists, and the description supplies the return rendering, the byte costs, and the truncation/redaction rules, which is most of what a 15-param query tool needs. Gaps are minor: no guidance on handling large result sets beyond pagination, and no note on what an empty match looks like.

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

Parameters5/5

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

With 100% schema coverage the baseline is 3, but the description goes beyond it with semantics the schema does not encode: array filters are "OR within one and AND across", `index` pulls full bodies for specific timeline events, headers cost "about 600 bytes a request", and the first line carries the match count and UTC anchor for aligning with backend logs.

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?

Opens with a specific resource and output shape: "The Jam's network requests as text, one per line" plus an inline template. That clearly separates it from getConsoleLogs or getEvents, but the description never names a sibling, so an agent must infer the boundary itself.

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

Usage Guidelines3/5

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

Usage is implied through filter recipes — "statusCode (['4xx','5xx'] for failures)", "method ([...] for writes)", "filter first" before bodies:'all', and format:'json' "for digging into one event or window" — but there is no explicit statement of when to pick this tool over getEvents or getUserEvents, and no stated exclusions.

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

All Jam toolsInspect
Try in Browser
Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources