Skip to main content
Glama
Sign In
codebase-memory-mcp

codebase-memory-mcp

by DeusData
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Local

ingest_traces

Destructive

Record observed runtime call paths or caller/callee pairs to distinguish executed routes from merely possible ones in a code knowledge graph; reports unresolved names per run label.

Instructions

Record observed runtime calls so the graph can distinguish observed from merely possible paths. Each trace is a call path (qualified names, adjacent = caller→callee) or a bare caller/callee pair. Stored per run label, keyed by qualified name (reindex-proof). Only pairs whose both endpoints are indexed symbols are stored; the response reports unresolved names (up to 10) so the producer can fix its naming.

Input Schema

NameRequiredDescriptionDefault
labelNoRun identity, e.g. 'pytest 2026-08-27' (default 'unlabeled'); counts accumulate per label and whole runs age out together
tracesYes
projectYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv0.12.1
    • addedInput schema / properties / label
      Added value: +{
+  "description": "Run identity, e.g. 'pytest 2026-08-27' (default 'unlabeled'); counts accumulate per label and whole runs age out together",
+  "type": "string"
+}
    • addedInput schema / properties / traces / items / properties / count / description
      Added value: +"How often this trace was observed (default 1)"
    • addedInput schema / properties / traces / items / properties / path
      Added value: +{
+  "description": "Call path of qualified names, caller before callee; 2..256 entries",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
  2. Changed1 schema field changedv0.10.4
    • changedOutput schema / (root)
      Previous value: -{
-  "additionalProperties": true,
-  "type": "object"
-}New value: +null
  3. Addedv0.10.0
  4. Removedv0.9.0
  5. First observedv1.0.0

TDQS

A3.9/5.0
Behavior4/5

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

Annotations already declare the write profile (readOnlyHint=false, destructiveHint=true, idempotentHint=false), and the description adds genuinely new behavior: storage is keyed per run label with whole runs aging out together, only pairs with both endpoints indexed are persisted, and unresolved names (up to 10) are reported back. It does not explicitly state the destructive/accumulating semantics in prose, but the per-label accumulation is consistent with idempotentHint=false, so there is no contradiction.

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?

Three sentences, purpose front-loaded, each carrying distinct information (goal, accepted formats, storage/reporting behavior). Dense but no filler; slightly compressed for the amount of behavior packed in.

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?

For a destructive mutation tool with no output schema, the description usefully covers what happens on write (partial acceptance, unresolved-name reporting, per-label accumulation). The main gap is the unexplained 'project' parameter and the absence of any explicit statement about permission or auth requirements.

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

Parameters3/5

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

Schema description coverage is only 33%, so the description must carry more than the baseline. It compensates for the undocumented caller/callee fields by explaining a trace may be a bare caller/callee pair and clarifies adjacency ('adjacent = caller→callee') and the per-run-label keying, but 'project' is never explained and 'count' semantics are left entirely to the schema.

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

Purpose5/5

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

States a specific verb and resource ('Record observed runtime calls') plus the goal that separates it from graph-construction siblings ('distinguish observed from merely possible paths'). An agent can tell this is the trace-ingestion write path, not index_repository or search_graph.

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?

The context of use (reporting runtime call paths so the graph can mark observed edges) is implied, and the trace formats are enumerated, but there is no explicit 'use this after running your test suite' guidance and no named alternative or when-not-to-use condition against siblings like index_repository.

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

Deploy Server

Other Tools