agents-deploy

Name: agents-deploy
Author: aws

589installs190forks

How to install agents-deploy?

aws/agent-toolkit-for-aws/agents-deploy

$npx -y skills add aws/agent-toolkit-for-aws --skill agents-deploy

Installs into the current project.

Summary

Use when deploying your agent to AWS, or when a deploy has failed. Handles pre-flight validation, CDK/IAM/quota error diagnosis, version management, rollback, and canary deployments. Triggers on: "deploy my agent", "agentcore deploy", "deploy failed", "CDK error", "rollback", "ca

Files · 1

View on GitHub

SKILL.md

1	# deploy
2
3	Deploy your AgentCore agent to AWS, or diagnose why a deploy failed.
4
5	## When to use
6
7	- You're ready to deploy and want to validate config first
8	- `agentcore deploy` failed with an error
9	- You want to preview what deploy will create without actually deploying
10	- You want to deploy to a specific target (staging, production)
11	- You need to roll back to a previous version, pin to a specific version, or set up canary deployments
12
13	## Input
14
15	`$ARGUMENTS` is optional:
16
17	```
18	/agents-deploy # interactive — pre-flight check or diagnose failure
19	/agents-deploy preflight # validate config and IAM before deploying
20	/agents-deploy diagnose # diagnose a failed deploy (paste error or read logs)
21	/agents-deploy preview # show what deploy will create without deploying
22	/agents-deploy rollback # roll back to a previous version
23	```
24
25	## Process
26
27	### Step 0: Verify CLI version
28
29	Run `agentcore --version`. This skill requires v0.9.0 or later. If the version is older, tell the developer to run `agentcore update` before proceeding.
30
31	### Step 1: Determine the situation
32
33	Read `agentcore/agentcore.json` and `agentcore/aws-targets.json` if they exist.
34
35	Ask (or infer from context):
36
37	> "Are you:
38	>
39	> 1. About to deploy and want to check everything first
40	> 2. Dealing with a failed deploy — what error did you see?
41	> 3. Needing to roll back or pin a specific version?"
42
43	If the developer needs versioning, rollback, or canary deployment, load [`references/versioning.md`](references/versioning.md) and follow its instructions.
44
45	---
46
47	## Path A: Pre-flight validation
48
49	Run these checks before `agentcore deploy`:
50
51	### Check 1: Validate config files
52
53	Show the developer this command to run:
54
55	```bash
56	agentcore validate
57	```
58
59	This catches malformed `agentcore.json` before CDK even starts.
60
61	### Check 2: Verify region alignment
62
63	The most common deploy failure is a region mismatch. Show the developer these commands to verify:
64
65	```bash
66	# Your configured AWS region
67	aws configure get region
68
69	# The region in your deployment target
70	cat agentcore/aws-targets.json
71
72	# The account you're actually authenticated as
73	aws sts get-caller-identity
74	```
75
76	The `region` in `aws-targets.json` must match your `aws configure` default region. The `account` must match the account ID from `sts get-caller-identity`.
77
78	### Check 3: Verify Bedrock model access
79
80	Show the developer this command to check enabled models in their region:
81
82	```bash
83	aws bedrock list-foundation-models --region $(aws configure get region) \
84	--query 'modelSummaries[?modelLifecycle.status==`ACTIVE`].modelId' \
85	--output table
86	```
87
88	Cross-region inference profile IDs use a geographic prefix (`us.`, `eu.`, `apac.`) or `global.` to control where inference runs. The CLI scaffolds `global.` by default (e.g., `global.anthropic.claude-sonnet-4-5-20250929-v1:0`), which routes to any commercial region. Geographic prefixes keep inference within that geography (e.g., `eu.` stays in EU regions). All prefixes require model access enabled in every destination region the profile covers. Check the Bedrock docs for which regions are included in each profile prefix.
89
90	### Check 4: Preview what will be deployed
91
92	```bash
93	agentcore deploy --dry-run
94	agentcore deploy --diff
95	```
96
97	`--dry-run` shows what resources will be created. `--diff` shows the CDK diff against what's currently deployed.
98
99	### Check 5: Verify IAM permissions
100
101	Show the developer the permissions needed and this verification command:
102
103	```bash
104	aws iam simulate-principal-policy \
105	--policy-source-arn $(aws sts get-caller-identity --query Arn --output text) \
106	--action-names iam:CreateRole \
107	--resource-arns "arn:aws:iam:::role/BedrockAgentCore*"
108	```
109
110	### Run the deploy
111
112	```bash
113	agentcore deploy -y # auto-confirm (alias: agentcore dp -y)
114	agentcore deploy -y -v # verbose — shows resource-level events
115	agentcore deploy --target staging -y # deploy to a specific target
116	```
117
118	Memory provisioning note: If your project includes memory, deploy takes 2–5 minutes longer while the memory resource becomes ACTIVE. This is normal — not an error. Check status:
119
120	```bash
121	agentcore status --type memory
122	```
123
124	---
125
126	## Path B: Diagnose a failed deploy
127
128	### Step B1: Read the error
129
130	If the developer pasted an error,

People who liked this also viewed

6 picks

Type

agents-deploy

589installs190forks

View on GitHub

How to install agents-deploy?

aws/agent-toolkit-for-aws/agents-deploy

$npx -y skills add aws/agent-toolkit-for-aws --skill agents-deploy

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# deploy
2
3	Deploy your AgentCore agent to AWS, or diagnose why a deploy failed.
4
5	## When to use
6
7	- You're ready to deploy and want to validate config first
8	- `agentcore deploy` failed with an error
9	- You want to preview what deploy will create without actually deploying
10	- You want to deploy to a specific target (staging, production)
11	- You need to roll back to a previous version, pin to a specific version, or set up canary deployments
12
13	## Input
14
15	`$ARGUMENTS` is optional:
16
17	```
18	/agents-deploy # interactive — pre-flight check or diagnose failure
19	/agents-deploy preflight # validate config and IAM before deploying
20	/agents-deploy diagnose # diagnose a failed deploy (paste error or read logs)
21	/agents-deploy preview # show what deploy will create without deploying
22	/agents-deploy rollback # roll back to a previous version
23	```
24
25	## Process
26
27	### Step 0: Verify CLI version
28
29	Run `agentcore --version`. This skill requires v0.9.0 or later. If the version is older, tell the developer to run `agentcore update` before proceeding.
30
31	### Step 1: Determine the situation
32
33	Read `agentcore/agentcore.json` and `agentcore/aws-targets.json` if they exist.
34
35	Ask (or infer from context):
36
37	> "Are you:
38	>
39	> 1. About to deploy and want to check everything first
40	> 2. Dealing with a failed deploy — what error did you see?
41	> 3. Needing to roll back or pin a specific version?"
42
43	If the developer needs versioning, rollback, or canary deployment, load [`references/versioning.md`](references/versioning.md) and follow its instructions.
44
45	---
46
47	## Path A: Pre-flight validation
48
49	Run these checks before `agentcore deploy`:
50
51	### Check 1: Validate config files
52
53	Show the developer this command to run:
54
55	```bash
56	agentcore validate
57	```
58
59	This catches malformed `agentcore.json` before CDK even starts.
60
61	### Check 2: Verify region alignment
62
63	The most common deploy failure is a region mismatch. Show the developer these commands to verify:
64
65	```bash
66	# Your configured AWS region
67	aws configure get region
68
69	# The region in your deployment target
70	cat agentcore/aws-targets.json
71
72	# The account you're actually authenticated as
73	aws sts get-caller-identity
74	```
75
76	The `region` in `aws-targets.json` must match your `aws configure` default region. The `account` must match the account ID from `sts get-caller-identity`.
77
78	### Check 3: Verify Bedrock model access
79
80	Show the developer this command to check enabled models in their region:
81
82	```bash
83	aws bedrock list-foundation-models --region $(aws configure get region) \
84	--query 'modelSummaries[?modelLifecycle.status==`ACTIVE`].modelId' \
85	--output table
86	```
87
88	Cross-region inference profile IDs use a geographic prefix (`us.`, `eu.`, `apac.`) or `global.` to control where inference runs. The CLI scaffolds `global.` by default (e.g., `global.anthropic.claude-sonnet-4-5-20250929-v1:0`), which routes to any commercial region. Geographic prefixes keep inference within that geography (e.g., `eu.` stays in EU regions). All prefixes require model access enabled in every destination region the profile covers. Check the Bedrock docs for which regions are included in each profile prefix.
89
90	### Check 4: Preview what will be deployed
91
92	```bash
93	agentcore deploy --dry-run
94	agentcore deploy --diff
95	```
96
97	`--dry-run` shows what resources will be created. `--diff` shows the CDK diff against what's currently deployed.
98
99	### Check 5: Verify IAM permissions
100
101	Show the developer the permissions needed and this verification command:
102
103	```bash
104	aws iam simulate-principal-policy \
105	--policy-source-arn $(aws sts get-caller-identity --query Arn --output text) \
106	--action-names iam:CreateRole \
107	--resource-arns "arn:aws:iam:::role/BedrockAgentCore*"
108	```
109
110	### Run the deploy
111
112	```bash
113	agentcore deploy -y # auto-confirm (alias: agentcore dp -y)
114	agentcore deploy -y -v # verbose — shows resource-level events
115	agentcore deploy --target staging -y # deploy to a specific target
116	```
117
118	Memory provisioning note: If your project includes memory, deploy takes 2–5 minutes longer while the memory resource becomes ACTIVE. This is normal — not an error. Check status:
119
120	```bash
121	agentcore status --type memory
122	```
123
124	---
125
126	## Path B: Diagnose a failed deploy
127
128	### Step B1: Read the error
129
130	If the developer pasted an error,

People who liked this also viewed

6 picks

Type