multiagent-orchestration changedmanaged-agents/multiagent-orchestration
Nearest release: v2.1.301, published 5 hours before this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+44added
Lines−55removed
From line
116
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits27to this page, all time
The whole hunk
from line 116, old and new numbered
/
from line 116
116116 -H "anthropic-version: 2023-06-01" \
117117 -H "anthropic-beta: managed-agents-2026-04-01" \
118118 -H "content-type: application/json" \
119 -d @- <<EOF
120 {
121 "name": "Engineering Lead",
122 "model": "claude-opus-5-5",
123 "system": "You coordinate engineering work. Delegate code review to the reviewer agent and test writing to the test agent.",
124 "tools": [
125 {
126 "type": "agent_toolset_20260401"
119 -d '{
120 "name": "Engineering Lead",
121 "model": "claude-opus-5-5",
122 "system": "You coordinate engineering work. Delegate code review to the reviewer agent and test writing to the test agent.",
123 "tools": [
124 {
125 "type": "agent_toolset_20260401"
126 }
127 ],
128 "multiagent": {
129 "type": "multiagent_20261001",
130 "subagents": {
131 "type": "enabled",
132 "predefined_agents": [
133 {"type": "agent", "id": "'"$REVIEWER_AGENT_ID"'"},
134 {"type": "agent", "id": "'"$TEST_WRITER_AGENT_ID"'"}
135 ]
136 }
127137 }
128 ],
129 "multiagent": {
130 "type": "multiagent_20261001",
131 "subagents": {
132 "type": "enabled",
133 "predefined_agents": [
134 {"type": "agent", "id": "$REVIEWER_AGENT_ID"},
135 {"type": "agent", "id": "$TEST_WRITER_AGENT_ID"}
136 ]
137 }
138 }
139 }
140 EOF
138 }'
141139 )
142140 ```
143141
from line 381
383381 -H "anthropic-version: 2023-06-01" \
384382 -H "anthropic-beta: managed-agents-2026-04-01" \
385383 -H "content-type: application/json" \
386 -d @- <<EOF
387 {
388 "agent": "$LEAD_AGENT_ID",
389 "environment_id": "$ENVIRONMENT_ID"
390 }
391 EOF
384 -d '{
385 "agent": "'"$LEAD_AGENT_ID"'",
386 "environment_id": "'"$ENVIRONMENT_ID"'"
387 }'
392388 )
393389 SESSION_ID=$(jq -r '.id' <<< "$session")
394390 ```
from line 466
470466
471467<CodeGroup defaultLanguage="CLI">
472468 ```bash cURL
473 research_agent_id=$(curl --fail-with-body -sS "$BASE/v1/agents" "${H[@]}" --data @- <<'EOF' | jq -er '.id'
474 {
469 research_agent_id=$(curl --fail-with-body -sS "$BASE/v1/agents" "${H[@]}" --data '{
475470 "name": "researcher",
476471 "model": "claude-haiku-5-5",
477472 "mcp_servers": [{"type": "url", "name": "github", "url": "https://api.githubcopilot.com/mcp/"}],
478473 "tools": [{"type": "mcp_toolset", "mcp_server_name": "github"}]
479 }
480 EOF
474 }' | jq -er '.id'
481475 )
482476
483 lead_agent_id=$(curl --fail-with-body -sS "$BASE/v1/agents" "${H[@]}" --data @- <<EOF | jq -er '.id'
484 {
477 lead_agent_id=$(curl --fail-with-body -sS "$BASE/v1/agents" "${H[@]}" --data '{
485478 "name": "lead",
486479 "model": "claude-opus-5-5",
487480 "tools": [{"type": "agent_toolset_20260401"}],
from line 482
489482 "type": "multiagent_20261001",
490483 "subagents": {
491484 "type": "enabled",
492 "predefined_agents": [{"type": "agent", "id": "$research_agent_id"}]
485 "predefined_agents": [{"type": "agent", "id": "'"$research_agent_id"'"}]
493486 }
494487 }
495 }
496 EOF
488 }' | jq -er '.id'
497489 )
498490 ```
499491
from line 782
790782
791783<CodeGroup>
792784 ```bash cURL
793 session_id=$(curl --fail-with-body -sS "$BASE/v1/sessions" "${H[@]}" --data @- <<EOF | jq -er '.id'
794 {
795 "agent": "$lead_agent_id",
796 "environment_id": "$environment_id",
797 "vault_ids": ["$vault_id"]
798 }
799 EOF
785 session_id=$(curl --fail-with-body -sS "$BASE/v1/sessions" "${H[@]}" --data '{
786 "agent": "'"$lead_agent_id"'",
787 "environment_id": "'"$environment_id"'",
788 "vault_ids": ["'"$vault_id"'"]
789 }' | jq -er '.id'
800790 )
801791 echo "$session_id"
802792 ```
from line 901
911901 -H "anthropic-version: 2023-06-01" \
912902 -H "anthropic-beta: managed-agents-2026-04-01" \
913903 -H "content-type: application/json" \
914 -d @- <<'EOF'
915 {
916 "name": "Contract Reviewer",
917 "model": "claude-opus-5-5",
918 "system": "You review contracts. When you're asked to review more than a few contracts, start a workflow run that reads them in parallel and combines the findings. Review one or two contracts yourself, without a run.",
919 "tools": [{"type": "agent_toolset_20260401"}],
920 "multiagent": {"type": "multiagent_20261001", "workflows": {"type": "enabled"}}
921 }
922 EOF
904 -d '{
905 "name": "Contract Reviewer",
906 "model": "claude-opus-5-5",
907 "system": "You review contracts. When you are asked to review more than a few contracts, start a workflow run that reads them in parallel and combines the findings. Review one or two contracts yourself, without a run.",
908 "tools": [{"type": "agent_toolset_20260401"}],
909 "multiagent": {"type": "multiagent_20261001", "workflows": {"type": "enabled"}}
910 }'
923911 ```
924912
925913 <CodeGroupItem>
from line 1105
11171105To adapt it, name the tasks in your agent's domain that call for a run, and the small tasks the agent should do itself. You can also guide how a run does the work, as in these examples:
11181106
11191107* **How to split the work:** "In a run, use one agent per contract. Then have a second agent review every contract, not only the ones where the first agent found something, and look for what the first agent missed. Anything that fails the review gets redone and reviewed again."
1120* **What to do when an agent fails:** "One agent failing must not fail the run. If an agent fails to read a contract, list that contract as not covered."
1108* **What to do when an agent fails:** "One agent failing must not fail the run. If an agent fails to read a contract, list that contract as not covered. When the run ends, tell me which contracts weren't covered."
11211109* **A time limit:** "Give the run a one-hour time limit."
1110* **Plan first:** "Before you start a run, describe its phases and how many agents each one runs. Then wait for my go-ahead." There's no approval step before a run starts, so nothing enforces this.
11221111
11231112Keep runs to the tasks that need them, because every agent in a run uses tokens.
11241113
No line in this hunk matches that.