Skip to main content
Glama

get_task

Read-onlyIdempotent

Retrieve task details with compact or full views, including status, comments, artifacts, dependencies, and VCS links. Use it to check task state and recent changes.

Instructions

Get task details: compact decision fields, active human_gate=true and status slug; descriptions and ordinary comment bodies are always complete. Comment IDs support reply/after; timestamps are UTC to the minute. Known automation comments are one-line summaries (include_auto=true restores their complete bodies). full=true restores non-capability REST fields. If status lookup is unavailable, status_id and an explicit status_lookup_error remain. Comments return the newest comments_limit entries (default 5, max 200) with pagination markers. since=RFC3339 (from task.updated_at) skips an unchanged task and older comments. In a stdio session a repeat read of the same card returns only what changed since that session last read it (unchanged card: stub with task_changed=false; unchanged description omitted with description_unchanged=true; comments already returned are dropped, counted in comments_already_shown); full=true always returns everything.

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": {
      "task_id": "<task_id>"
    },
    "name": "get_task"
  }
}

Input Schema

NameRequiredDescriptionDefault
fullNoReturn the task (full description, non-capability fields, whole inline comments) instead of the compact view (default false).
sinceNoRFC3339 timestamp from an earlier call in this session (e.g. that response's task.updated_at, or your own last-call time). With it: include_comments returns only comments created after it, and the full task body (description and the rest) is included only if the task's own updated_at is after it too — the response then carries task_changed=true/false so you always know which you got, and a trimmed stub (id/status/assignee_name/updated_at in compact reads; status_id plus status_lookup_error if lookup fails, or the original status_id and assignee fields with full=true) instead of the full task when nothing changed, to avoid re-paying for the same ~5k tokens on a repeat call. Omit for the unfiltered default: full task body, but the comment tail still obeys comments_limit (default 5, max 200) — a longer thread needs list_comments paging.
task_idYesTask ID (full UUID or 6–12 char hex short-ID prefix).
include_autoNoReturn complete automation bodies in compact mode (ordinary comments are always complete).
comments_limitNoHow many of the NEWEST comments to include when include_comments=true (default 5, max 200). The tail is where an acting agent looks first; raise it when you need more of it. The API caps any single request at 200 — a longer thread is only readable end-to-end by paging list_comments (order=desc, page=N), never by one call here.
include_commentsNoInclude the task's most recent comments (see comments_limit for how many; the response carries comments_total_count and comments_has_more so a longer thread is never silently hidden).
include_artifactsNoInclude artifacts.
include_vcs_linksNoInclude linked PRs/MRs/commits/branches (id, provider, link_type, external_id, url, status, created_at, plus delivery{implementation_sha, ci_verdict, merged_sha, deployed, acceptance} and a derived is_shipped) — use this instead of a raw REST call to diagnose a misclassified or stuck-status link.
include_dependenciesNoInclude dependencies.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv0.1.6
    • changedInput schema / properties / include_vcs_links / description
      Previous value: -"Include linked PRs/MRs/commits/branches (id, provider, link_type, external_id, url, status, created_at) — use this instead of a raw REST call to diagnose a misclassified or stuck-status link."New value: +"Include linked PRs/MRs/commits/branches (id, provider, link_type, external_id, url, status, created_at, plus delivery{implementation_sha, ci_verdict, merged_sha, deployed, acceptance} and a derived is_shipped) — use this instead of a raw REST call to diagnose a misclassified or stuck-status link."
  2. First observedv0.1.1

TDQS

A4.5/5.0
Behavior5/5

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

Annotations already declare a safe idempotent read, and the description layers on behavior those annotations cannot express: stdio sessions return only deltas on repeat reads (task_changed=false stubs, description_unchanged=true, comments_already_shown counts), automation comments are collapsed to one-liners, and status_id with an explicit status_lookup_error survives a failed lookup. These are exactly the surprises that would otherwise mislead an agent about missing data.

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?

Every sentence carries substantive information, but it arrives as one dense run-on paragraph mixing purpose, caching, truncation, and parameter semantics with no headings or ordering beyond the first clause. Defaults and max values are restated in both description and schema, adding length without new meaning.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With nine parameters, no output schema, and non-obvious session-delta semantics, the definition supplies what the structured data cannot: delta behavior, comment truncation and its escape hatches, since semantics, and the status-lookup failure path. An agent can call this correctly and interpret the response without further context.

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, but the description earns more by adding response-side meaning: comments_limit governs the newest N with pagination markers, comments_total_count and comments_has_more prevent silent truncation, and since also filters comments and carries task_changed. Timestamps being UTC-to-the-minute and comment IDs supporting reply/after are details absent from 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?

Opens with a specific verb and resource ("Get task details") and immediately defines the scope: compact decision fields by default, with full=true restoring non-capability REST fields. It also distinguishes the read from the sibling path for long threads by naming list_comments paging as the alternative, so an agent can place it 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 Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete when-to-use triggers for several switches: since=RFC3339 to skip an unchanged task, include_auto=true only when complete automation bodies are needed, full=true for non-capability fields, and list_comments when a thread exceeds the 200-comment cap. It stops short of stating when to prefer this over the closest sibling get_task_context, so it is clear context rather than a full routing rule.

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