Skip to main content
Glama

control_execution

Destructive

Pause, resume, steer, stop, abort, purge, or hand off an agent run and its workers/tasks to regain human control during execution.

Instructions

Steer a heads-down run when the human asks. pause / resume: the whole run, or only task_ids (paused work starts nothing new; running workers finish their turn). steer: a directive (message) to targets (agents, teams or tasks; default every agent), which workers must acknowledge. abort: stop the run and its workers, keeping the worktrees, branches and log. purge: delete a finished (completed, failed or aborted) run's worktrees, branches and log; needs confirm: true, which only the human's agreement justifies, and is audited. stop_agent: stop the live workers named in targets (the human asked, in plain words or from the dashboard): the runner stops each worker, keeps its branch and worktree and blocks its task, and the watcher wakes you with two choices: rework around the stopped agent with resolve_escalation, or write a handoff. An external worker is told to stop through steer instead. handoff: pause the run (if live) and write executions//HANDOFF.md from the blackboard (contract summary, goal, every task with its state, owner, branch and last note, stopped agents, open escalations, budget with provenance split, resume commands); returns its path and the document. Back to the table: pause with to_table: true (recorded in the log as a mode change), run the format with execution_status(view: "context"), amend_contract(from_session), then resume with to_table: true.

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": {
      "execution_id": "<execution_id>",
      "action": "<action>"
    },
    "name": "control_execution"
  }
}

Input Schema

NameRequiredDescriptionDefault
actionYespause, resume, steer, abort, purge, stop_agent or handoff.
reasonNoWhy (recorded on the event).
confirmNopurge: true once the human agreed to delete the run's worktrees, branches and log.
messageNosteer: the directive, in plain words.
sessionNoWith to_table: the at-the-table session id, recorded with the mode change.
targetsNosteer: agent, team or task ids (default every agent of the run). stop_agent: the agents to stop (required).
task_idsNopause / resume: only these tasks (default the whole run).
to_tableNopause: going back to the table; resume: coming back. Records the mode change in the log.
execution_idYesThe heads-down run (exe-…), as execute or execution_status returned it.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv2.0.0

TDQS

A4.6/5.0
Behavior5/5

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

Annotations already declare destructiveHint=true, readOnlyHint=false and idempotentHint=false, but the description goes well past them: it states exactly what abort preserves versus what purge deletes, that purge requires confirm:true justified only by human agreement and is audited, that paused work starts nothing new while running workers finish their turn, and that stop_agent blocks the task and wakes the agent with two options. It even documents handoff's return value (path and document). This is unusually rich behavioral disclosure for a destructive tool.

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 front-loaded with the governing condition ('when the human asks') and wastes few words, but it is one dense paragraph of chained parentheticals that is hard to scan for a specific action. A per-action structure would deliver the same content far more legibly. Length is justified by seven distinct actions, but the formatting is not.

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 nine-parameter destructive tool with no output schema, the description covers the critical ground: what each action mutates, what is preserved, the audit and human-consent requirements, and the return of handoff. It stops short of saying what the lighter actions return or how failures surface, which an agent calling pause or steer would still want.

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?

Schema coverage is 100%, so the baseline is 3; the description earns a bump by adding meaning the schema does not carry — the default for targets on steer (every agent of the run), the default for task_ids (the whole run), the provenance requirement behind confirm, and that to_table is recorded as a mode change and pairs with session. A few params (reason, message) are only echoed from the schema rather than enriched.

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 names each of the seven actions and what each does at a concrete verb+resource level ('abort: stop the run and its workers, keeping the worktrees, branches and log'; 'handoff: pause the run ... and write executions/<exe>/HANDOFF.md'). It distinguishes itself from siblings by naming resolve_escalation, execution_status and amend_contract where those are the right tool instead. An agent can tell which action to pick without opening the schema.

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?

Every action carries a when-to-use condition: 'when the human asks', 'the human asked, in plain words or from the dashboard', and it explicitly routes a case elsewhere ('An external worker is told to stop through steer instead'). It also spells out a multi-tool sequence for returning to the table (pause with to_table, execution_status(view: "context"), amend_contract(from_session), resume with to_table). Nothing about selecting this tool is left to inference.

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