Skip to main content
Glama

job_start

Destructive

Start a local or SSH command as a tracked job, with idempotent retries via a trace ID to prevent duplicate execution.

Instructions

Start one tracked local or SSH command. Pass remote shell commands verbatim without adding escapes. For a RunBeacon prompt trace, reuse requestTraceId on every retry so the server returns the original job instead of executing twice. When the user requests the default SSH server, set useDefaultCredential=true and call this tool directly without listing profiles first.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
cwdNoLocal working directory.
envNoLocal environment overrides. Values are never persisted.
argsNoLocal command arguments. For SSH, include arguments in command.
labelNo
shellNoUse a local shell.
targetNo
adapterNogeneric
commandYesLocal command or complete remote shell command.
metadataNoIn-memory caller metadata. It is not persisted unless RUNBEACON_PERSIST_METADATA=true, and sensitive-key values are redacted when persistence is enabled.
timeoutMsNo
outputPolicyNo
executionModeNoPrefer the durable runner, force direct SSH, or require the runner path.auto
idempotencyKeyNoStable caller-provided key that returns the existing job instead of launching a duplicate after a retry.
requestTraceIdNoOpaque request UUID supplied by the RunBeacon prompt hook. Reuse it unchanged; a second start with the same trace returns the first job.
requireDurableNoFail instead of falling back to direct SSH when the durable runner is unavailable.
progressPatternNoOptional RE2-compatible regex (no backreferences or lookbehind); capture group 1 must contain a finite percentage.
credentialProfileNoSaved SSH profile. If omitted, a unique profile matching target.host and target.username is selected automatically when no inline authentication is supplied.
requestReceivedAtNoISO timestamp supplied by the RunBeacon prompt hook for end-to-end latency measurement.
eventSubscriptionsNo
useDefaultCredentialNoFast path for an explicitly requested default SSH server. Set true directly without calling credential_profile_list first. Leave false for local commands.
Behavior4/5

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

Annotations already indicate destructive/write behavior, and the description does not contradict them. It adds useful behavioral context about idempotent retry via requestTraceId, verbatim command pass-through, and the default-credential fast path, which aids the agent beyond the safety flags.

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

Conciseness5/5

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

Three sentences, each carrying actionable information, front-loaded with the primary purpose. There is no filler or repetition of schema details.

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?

Given the tool's complexity (20 params, nested target objects, no output schema), the description covers the most critical operational scenarios: retry idempotency and the default SSH fast path. It doesn't enumerate every parameter, but the schema handles those; the description provides the decision context an agent needs.

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?

With schema coverage at ~70%, the description adds meaning to key parameters: requestTraceId reuse semantics, useDefaultCredential direct path, and remote command escaping behavior. While not covering all 20 parameters, it enriches those that are most likely to cause misuse.

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?

The description opens with 'Start one tracked local or SSH command,' which is a specific verb+resource that distinguishes it from sibling job tools like job_cancel or job_list. It clearly communicates the core action without ambiguity.

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?

The description gives explicit directives: pass remote commands verbatim, reuse requestTraceId on retries, and set useDefaultCredential=true to skip profile listing. It names a specific alternative (credential_profile_list) and explicitly says when not to use it.

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

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Liyuchen0118/RunBeacon'

If you have feedback or need assistance with the MCP directory API, please join our Discord server