coordinating-multi-space-devops-agent

Name: coordinating-multi-space-devops-agent
Author: aws

249installs190forks

How to install coordinating-multi-space-devops-agent?

aws/agent-toolkit-for-aws/coordinating-multi-space-devops-agent

$npx -y skills add aws/agent-toolkit-for-aws --skill coordinating-multi-space-devops-agent

Installs into the current project.

Summary

Coordinate the AWS DevOps Agent across multiple AgentSpaces from one Claude Code session — route questions to the right space (prod vs staging vs knowledge), query several spaces in parallel and synthesize, or compare findings across accounts. Use whenever the user has more than

Files · 1

View on GitHub

SKILL.md

1	# Querying multiple AgentSpaces
2
3	## Pre-flight
4
5	If `aws_devops_agent__list_agent_spaces` is not in your available tools, the remote MCP server is not connected. Tell the user to ask "help me set up the AWS DevOps Agent" so the `setup-devops-agent` skill auto-loads.
6
7	## Prerequisite: SigV4 auth required
8
9	Multi-space routing requires SigV4 authentication — Bearer tokens are scoped to a single AgentSpace and cannot route to other spaces.
10
11	Many real teams run more than one AgentSpace — typically a production space, a staging space, and a dedicated "knowledge" space that holds runbooks shared across accounts. Each space has its own set of associated AWS accounts, runbooks, and history.
12
13	This skill is the routing brain. Use it when the user has multiple spaces configured, or when a question genuinely spans accounts.
14
15	## Discovering spaces
16
17	```
18	aws_devops_agent__list_agent_spaces()
19	→ {"agentSpaces": [{"agentSpaceId": "as-abc123", "name": "prod"}, ...]}
20	```
21
22	If only one space is returned, this skill doesn't apply — use `chatting-with-aws-devops-agent` or `investigating-incidents-with-aws-devops-agent` directly (no `agent_space_id` needed).
23
24	If more than one is returned, decide whether the user's question is:
25
26	\| Question shape \| Strategy \|
27	\|---------------\|----------\|
28	\| Scoped to one environment ("prod is broken") \| Single space — pick the matching one \|
29	\| Spans environments ("compare prod vs staging") \| Parallel — query each, synthesize \|
30	\| Generic knowledge ("what runbooks do we have for ECS?") \| Route to the knowledge space if one is named that way \|
31	\| Ambiguous ("our service is slow") \| Ask the user which environment, don't guess \|
32
33	## Per-session routing memory
34
35	If the user has a routing guide stored locally (e.g. `.claude/aws-agents-for-devsecops.md`, `AGENTS.md`, or per-project notes), read it once at the start of the session and use it as the routing table for the rest of the conversation. Format expected:
36
37	```markdown
38	\| Space \| AWS Profile \| Agent Space ID \| Purpose \|
39	\|-------\|-------------\|----------------\|---------\|
40	\| prod \| acme-prod \| as-abc123 \| Production incidents, customer-facing services \|
41	\| stage \| acme-stage \| as-def456 \| Pre-prod validation, integration testing \|
42	\| kb \| acme-shared \| as-ghi789 \| Shared runbooks, cross-account knowledge \|
43	```
44
45	If no guide exists, run discovery:
46
47	1. `aws_devops_agent__list_agent_spaces()` → get all spaces.
48	2. For each space: `aws_devops_agent__chat(message="Summarize the AWS accounts, services, and runbooks you have access to.", agent_space_id="<SPACE_ID>")` → get a one-paragraph summary.
49	3. Offer to write the routing guide to the project (e.g. `.claude/aws-agents-for-devsecops.md`, `AGENTS.md`, or per-project notes) so future sessions skip discovery.
50
51	## Pattern A — Parallel queries, one synthesized answer
52
53	Use when the user wants a comparison: "compare prod and staging error rates", "is this issue happening in both accounts?", "audit costs across all our environments".
54
55	```
56	# 1. Query each space in parallel with environment-specific context
57	aws_devops_agent__chat(message="<question> \| env=prod \| <prod IaC context>", agent_space_id="PROD_ID")
58	→ {"executionId": "...", "answer": "..."}
59
60	aws_devops_agent__chat(message="<question> \| env=stage \| <stage IaC context>", agent_space_id="STAGE_ID")
61	→ {"executionId": "...", "answer": "..."}
62
63	# 2. Synthesize locally — present a side-by-side summary, not two separate dumps
64	```
65
66	Don't just paste both responses. Read both, identify what's the same vs. different, and tell the user the delta — that's the value.
67
68	## Pattern B — Knowledge lookup, then per-space action
69
70	Use when one space holds runbooks/knowledge that informs work in another space.
71
72	```
73	# 1. Ask the knowledge space first
74	aws_devops_agent__chat(
75	message="What's our standard runbook for ECS 503 errors?",
76	agent_space_id="KB_ID"
77	)
78	→ {"answer": "<runbook text>"}
79
80	# 2. Apply that runbook in the target environment
81	aws_devops_agent__investigate(
82	title="ECS 503 errors on checkout-service. [Runbook from knowledge space] <runbook text> [Local context] ...",
83	agent_space_id="PROD_ID",
84	priority="HIGH"
85	)
86	```
87
88	The DevOps Agent doesn't share state between spaces — you bridge it by quoting the knowledge space's response into the investigation's `title`.
89
90	## Pattern C — Targeted single-space query
91
92	Use when the user explicitly names a space or environment.
93
94	```
95	# Pick the matching agentSpaceId from your rout

People who liked this also viewed

6 picks

Type

coordinating-multi-space-devops-agent

249installs190forks

View on GitHub

How to install coordinating-multi-space-devops-agent?

aws/agent-toolkit-for-aws/coordinating-multi-space-devops-agent

$npx -y skills add aws/agent-toolkit-for-aws --skill coordinating-multi-space-devops-agent

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Querying multiple AgentSpaces
2
3	## Pre-flight
4
5	If `aws_devops_agent__list_agent_spaces` is not in your available tools, the remote MCP server is not connected. Tell the user to ask "help me set up the AWS DevOps Agent" so the `setup-devops-agent` skill auto-loads.
6
7	## Prerequisite: SigV4 auth required
8
9	Multi-space routing requires SigV4 authentication — Bearer tokens are scoped to a single AgentSpace and cannot route to other spaces.
10
11	Many real teams run more than one AgentSpace — typically a production space, a staging space, and a dedicated "knowledge" space that holds runbooks shared across accounts. Each space has its own set of associated AWS accounts, runbooks, and history.
12
13	This skill is the routing brain. Use it when the user has multiple spaces configured, or when a question genuinely spans accounts.
14
15	## Discovering spaces
16
17	```
18	aws_devops_agent__list_agent_spaces()
19	→ {"agentSpaces": [{"agentSpaceId": "as-abc123", "name": "prod"}, ...]}
20	```
21
22	If only one space is returned, this skill doesn't apply — use `chatting-with-aws-devops-agent` or `investigating-incidents-with-aws-devops-agent` directly (no `agent_space_id` needed).
23
24	If more than one is returned, decide whether the user's question is:
25
26	\| Question shape \| Strategy \|
27	\|---------------\|----------\|
28	\| Scoped to one environment ("prod is broken") \| Single space — pick the matching one \|
29	\| Spans environments ("compare prod vs staging") \| Parallel — query each, synthesize \|
30	\| Generic knowledge ("what runbooks do we have for ECS?") \| Route to the knowledge space if one is named that way \|
31	\| Ambiguous ("our service is slow") \| Ask the user which environment, don't guess \|
32
33	## Per-session routing memory
34
35	If the user has a routing guide stored locally (e.g. `.claude/aws-agents-for-devsecops.md`, `AGENTS.md`, or per-project notes), read it once at the start of the session and use it as the routing table for the rest of the conversation. Format expected:
36
37	```markdown
38	\| Space \| AWS Profile \| Agent Space ID \| Purpose \|
39	\|-------\|-------------\|----------------\|---------\|
40	\| prod \| acme-prod \| as-abc123 \| Production incidents, customer-facing services \|
41	\| stage \| acme-stage \| as-def456 \| Pre-prod validation, integration testing \|
42	\| kb \| acme-shared \| as-ghi789 \| Shared runbooks, cross-account knowledge \|
43	```
44
45	If no guide exists, run discovery:
46
47	1. `aws_devops_agent__list_agent_spaces()` → get all spaces.
48	2. For each space: `aws_devops_agent__chat(message="Summarize the AWS accounts, services, and runbooks you have access to.", agent_space_id="<SPACE_ID>")` → get a one-paragraph summary.
49	3. Offer to write the routing guide to the project (e.g. `.claude/aws-agents-for-devsecops.md`, `AGENTS.md`, or per-project notes) so future sessions skip discovery.
50
51	## Pattern A — Parallel queries, one synthesized answer
52
53	Use when the user wants a comparison: "compare prod and staging error rates", "is this issue happening in both accounts?", "audit costs across all our environments".
54
55	```
56	# 1. Query each space in parallel with environment-specific context
57	aws_devops_agent__chat(message="<question> \| env=prod \| <prod IaC context>", agent_space_id="PROD_ID")
58	→ {"executionId": "...", "answer": "..."}
59
60	aws_devops_agent__chat(message="<question> \| env=stage \| <stage IaC context>", agent_space_id="STAGE_ID")
61	→ {"executionId": "...", "answer": "..."}
62
63	# 2. Synthesize locally — present a side-by-side summary, not two separate dumps
64	```
65
66	Don't just paste both responses. Read both, identify what's the same vs. different, and tell the user the delta — that's the value.
67
68	## Pattern B — Knowledge lookup, then per-space action
69
70	Use when one space holds runbooks/knowledge that informs work in another space.
71
72	```
73	# 1. Ask the knowledge space first
74	aws_devops_agent__chat(
75	message="What's our standard runbook for ECS 503 errors?",
76	agent_space_id="KB_ID"
77	)
78	→ {"answer": "<runbook text>"}
79
80	# 2. Apply that runbook in the target environment
81	aws_devops_agent__investigate(
82	title="ECS 503 errors on checkout-service. [Runbook from knowledge space] <runbook text> [Local context] ...",
83	agent_space_id="PROD_ID",
84	priority="HIGH"
85	)
86	```
87
88	The DevOps Agent doesn't share state between spaces — you bridge it by quoting the knowledge space's response into the investigation's `title`.
89
90	## Pattern C — Targeted single-space query
91
92	Use when the user explicitly names a space or environment.
93
94	```
95	# Pick the matching agentSpaceId from your rout

People who liked this also viewed

6 picks

Type