backscroll
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_outputA | Full-text search over recorded terminal commands, their outputs, and user notes — finds e.g. every command that ever printed 'connection refused'. Returns plain text, one block per match: id, time, cwd, exit code, the command line, and a snippet around the match (secrets are masked by default). Use this to FIND which command said something; use list_commands to browse recent history without a search term, and get_output to read a full output once you have an id. No matches returns an empty result, not an error. Read-only: never re-runs anything. |
| list_commandsA | List the user's recent terminal commands (most recent first) as plain text: one entry per command with id, time, cwd, exit code, duration and output size — command lines only, no output text. Supports the same filters as search_output, e.g. exit='fail' for recent failures or cwd='.' for this project only. Use this to browse history; use search_output when looking for specific text, and get_output to read what a command actually printed. Read-only: never re-runs anything. |
| get_outputA | Get the full recorded output of one terminal command, plus its command line, exit code, cwd and timing, as plain text (ANSI escapes stripped; secrets masked by default). id: a command id from search_output/list_commands, or negative for relative addressing (-1 = the user's most recent command, -2 = the one before it). Outputs larger than max_bytes return the head and tail around a '[... N bytes omitted ...]' marker — call again with a larger max_bytes for more. Errors with 'not found' if the id doesn't exist. Read-only: reads the recording, never re-runs the command. |
| diff_outputA | Unified diff (diff -u style plain text) between the stored outputs of two runs. With only 'id', diffs that command against the most recent EARLIER run of the exact same command line — 'what changed since it last ran?'; errors if no earlier identical command line exists. With 'other', diffs the two given commands (other = older side). Identical outputs return a note saying so. Secrets are masked by default. Read-only: diffs recordings, never re-runs anything. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
Each tool has a clearly distinct purpose: search_output finds commands by text content, list_commands browses recent command metadata, get_output retrieves full output of a specific command, and diff_output compares outputs of two runs. No overlap in functionality.
All tool names follow a consistent verb_noun pattern (search_output, list_commands, get_output, diff_output) using lowercase with underscores. The style is uniform and predictable.
Four tools is well-scoped for the domain of querying terminal command recordings. Each tool covers a core operation (search, list, retrieve, diff) without unnecessary bloat or missing essentials.
The tools cover the main read-only use cases for terminal history: searching, browsing, viewing full output, and diffing. Minor gaps exist, such as lacking a tool to manage notes or delete recordings, but these are beyond the stated read-only scope.