Skip to main content
Glama

session_recall

Search a session transcript for exact text, including context compaction summarized away, and read specific messages in full from the live, still-running session.

Instructions

Get back exact text from one session, including what context compaction summarized away.

session_id: the session to search (the post-compaction checkpoint names it). query: words to find, e.g. an error string, file name or decision. source: claude | codex message: read message [mN] in full instead of searching.

Reads the transcript on disk directly, so it sees the current, still-running session.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "session_id": "<session_id>"
    },
    "name": "session_recall"
  }
}

Input Schema

NameRequiredDescriptionDefault
queryNo
sourceNoclaude
messageNo
max_hitsNo
session_idYes

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.4.0

TDQS

A3.7/5.0
Behavior4/5

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

With no annotations, the description carries the full burden, and it does disclose a non-obvious trait: it reads the transcript on disk directly, so it sees the current, still-running session and can surface content compaction removed. It omits failure modes (session not found, cross-machine/remote sessions) and any cost or size caveats, which a mutation-free read tool could reasonably leave out.

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?

Purpose is front-loaded in the first sentence, followed by a compact per-parameter list and one closing behavioral note — no filler. The terse "source: claude | codex" line is slightly more cryptic than needed but earns its place by supplying the allowed values.

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 shape need not be explained, and the description covers the tool's distinctive behavior (post-compaction recall, live disk read) plus four of five parameters. The gap is max_hits and what happens on a missing or ambiguous session_id — small omissions for a read-only retrieval tool but not zero.

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 coverage is 0%, so the description must compensate: it documents session_id (the post-compaction checkpoint names it), query with concrete examples, source values "claude | codex" (not encoded as an enum in the schema), and the message-mode flag. max_hits, one of five parameters, is never mentioned, leaving hit-count/truncation behavior undefined.

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?

"Get back exact text from one session" states a specific verb and resource, and "including what context compaction summarized away" carves out a distinct capability (verbatim transcript retrieval). It does not name or contrast with siblings such as get_session or search_history, so the agent must infer the boundary, but the purpose itself is unambiguous.

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 description gives intra-tool routing — use "query" to search by words, or "message" to read message [mN] in full instead of searching — which is clear guidance for choosing a mode. It offers no when-to-use versus sibling tools (get_session, search_history, search_graph) and no prerequisites, so usage is only implied at the tool-selection level.

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