post-deploy-qa

Name: post-deploy-qa
Author: pavel-molyanov

241stars75forks

How to install post-deploy-qa?

pavel-molyanov/molyanov-ai-dev/post-deploy-qa

$npx -y skills add pavel-molyanov/molyanov-ai-dev --skill post-deploy-qa

Installs into the current project.

Summary

Post-deploy verification: execute AVP from tech-spec on live environment, verify all acceptance criteria (user-spec + tech-spec), pick up deferred criteria from pre-deploy QA report. Uses MCP tools (Telegram MCP, Playwright, curl, bash). Use when: "пост-деплой проверка", "post-de

Files · 1

View on GitHub

SKILL.md

1	<!-- Generated by sync-to-codex v1. Do not edit directly. -->
2
3	# Post-deploy QA
4
5	## Input Requirements
6
7	Read before starting:
8	- `user-spec.md` — acceptance criteria
9	- `tech-spec.md` — Agent Verification Plan (AVP) section + technical acceptance criteria
10	- `decisions.md` — deviations from plan (if exists)
11	- Pre-deploy QA report (`logs/working/qa-report.json`) — check `deferredToPostDeploy` section for criteria that could not be verified pre-deploy
12	- Confirmation that deploy is complete and environment is live
13
14	If tech-spec has no AVP section — still proceed with acceptance criteria verification.
15
16	## Verification Methodology
17
18	Two verification directions (both required):
19
20	### 1. Agent Verification Plan (AVP)
21
22	Execute each step from AVP section in tech-spec:
23
24	1. Read AVP section — it lists verification steps with expected outcomes and MCP tools to use
25	2. For each step:
26	- Use the specified MCP tool (Telegram MCP, Playwright, curl, bash, etc.)
27	- Perform the described action on live environment
28	- Compare result with expected outcome
29	- Record: tool used, step performed, result
30	3. If MCP tool is unavailable — mark step as `not_verifiable`, continue with remaining steps
31
32	Checkpoint: All AVP steps executed or marked `not_verifiable`. Proceed to acceptance criteria.
33
34	### 2. Acceptance Criteria Verification
35
36	Verify all acceptance criteria from user-spec and tech-spec on live environment. This catches criteria that pre-deploy QA could not verify without a live system.
37
38	1. Read pre-deploy QA report (`logs/working/qa-report.json`) — check `deferredToPostDeploy` section
39	2. For each deferred criterion — follow the verification steps specified in the pre-deploy report
40	3. Also re-check all acceptance criteria from user-spec and tech-spec against live behavior
41	4. For each criterion:
42	- passed — verified on live environment, evidence provided
43	- failed — live behavior does not meet the criterion
44	- blocked — cannot verify due to external conditions (no data, third-party service down). Provide a concrete manual verification plan for the user: what to check, when, how
45
46	A `blocked` criterion requires user follow-up before the feature can count as fully verified.
47
48	Checkpoint: All acceptance criteria verified, or marked `blocked` with manual verification plan.
49
50	## Severity Classification
51
52	- critical — verification step failed, live functionality broken, data integrity at risk
53	- major — works but with significant issues visible in production
54	- minor — cosmetic, non-functional discrepancies
55
56	## Output Format
57
58	Return findings as JSON. Reason: orchestrator parses this to decide pass/fail and log findings.
59
60	```json
61	{
62	"status": "passed \| failed",
63	"summary": {
64	"totalSteps": 0,
65	"passed": 0,
66	"failed": 0,
67	"blocked": 0,
68	"notVerifiable": 0,
69	"criticals": 0,
70	"majors": 0,
71	"minors": 0
72	},
73	"agentVerification": [
74	{
75	"step": "Send /start to bot",
76	"tool": "telegram_mcp",
77	"status": "passed \| failed \| not_verifiable",
78	"details": "Bot responded with welcome message"
79	}
80	],
81	"acceptanceCriteria": [
82	{
83	"id": "US-5",
84	"criterion": "Titles generated with correct declensions",
85	"source": "user-spec \| tech-spec \| deferred-from-pre-deploy",
86	"status": "passed \| failed \| blocked",
87	"evidence": "Checked live output, generated title uses correct declension/grammar",
88	"manualVerificationPlan": "Only if blocked — what the user should check, when, how"
89	}
90	],
91	"findings": [
92	{
93	"severity": "critical \| major \| minor",
94	"title": "Bot does not respond to /start",
95	"expected": "Welcome message within 3 seconds",
96	"actual": "No response after 10 seconds",
97	"reproduction": "Steps to reproduce..."
98	}
99	]
100	}
101	```
102
103	Status decision: `passed` if zero criticals, `failed` if one or more criticals.
104
105	## Guidelines
106
107	- Every finding includes concrete reproduction: steps, expected vs actual, tool output.
108	- If all MCP tools are unavailable — report that automated verification is not possible, suggest manual steps.
109	- Empty findings array = clean verification.
110
111	## Final Check
112
113	Before finishing, verify:
114	- [ ] All AVP steps executed or marked `not_verifiable`
115	- [ ] All deferred criteria from pre-deploy QA report addressed
116	- [ ] All acceptance criteria from user-spec and tech-spec verified on live environment
117	- [ ] Every `blocked` criterion has a manual verification plan (what to check, when, how)
118	- [ ] Output JSON is vali

People who liked this also viewed

6 picks

Type

post-deploy-qa

241stars75forks

View on GitHub

How to install post-deploy-qa?

pavel-molyanov/molyanov-ai-dev/post-deploy-qa

$npx -y skills add pavel-molyanov/molyanov-ai-dev --skill post-deploy-qa

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	<!-- Generated by sync-to-codex v1. Do not edit directly. -->
2
3	# Post-deploy QA
4
5	## Input Requirements
6
7	Read before starting:
8	- `user-spec.md` — acceptance criteria
9	- `tech-spec.md` — Agent Verification Plan (AVP) section + technical acceptance criteria
10	- `decisions.md` — deviations from plan (if exists)
11	- Pre-deploy QA report (`logs/working/qa-report.json`) — check `deferredToPostDeploy` section for criteria that could not be verified pre-deploy
12	- Confirmation that deploy is complete and environment is live
13
14	If tech-spec has no AVP section — still proceed with acceptance criteria verification.
15
16	## Verification Methodology
17
18	Two verification directions (both required):
19
20	### 1. Agent Verification Plan (AVP)
21
22	Execute each step from AVP section in tech-spec:
23
24	1. Read AVP section — it lists verification steps with expected outcomes and MCP tools to use
25	2. For each step:
26	- Use the specified MCP tool (Telegram MCP, Playwright, curl, bash, etc.)
27	- Perform the described action on live environment
28	- Compare result with expected outcome
29	- Record: tool used, step performed, result
30	3. If MCP tool is unavailable — mark step as `not_verifiable`, continue with remaining steps
31
32	Checkpoint: All AVP steps executed or marked `not_verifiable`. Proceed to acceptance criteria.
33
34	### 2. Acceptance Criteria Verification
35
36	Verify all acceptance criteria from user-spec and tech-spec on live environment. This catches criteria that pre-deploy QA could not verify without a live system.
37
38	1. Read pre-deploy QA report (`logs/working/qa-report.json`) — check `deferredToPostDeploy` section
39	2. For each deferred criterion — follow the verification steps specified in the pre-deploy report
40	3. Also re-check all acceptance criteria from user-spec and tech-spec against live behavior
41	4. For each criterion:
42	- passed — verified on live environment, evidence provided
43	- failed — live behavior does not meet the criterion
44	- blocked — cannot verify due to external conditions (no data, third-party service down). Provide a concrete manual verification plan for the user: what to check, when, how
45
46	A `blocked` criterion requires user follow-up before the feature can count as fully verified.
47
48	Checkpoint: All acceptance criteria verified, or marked `blocked` with manual verification plan.
49
50	## Severity Classification
51
52	- critical — verification step failed, live functionality broken, data integrity at risk
53	- major — works but with significant issues visible in production
54	- minor — cosmetic, non-functional discrepancies
55
56	## Output Format
57
58	Return findings as JSON. Reason: orchestrator parses this to decide pass/fail and log findings.
59
60	```json
61	{
62	"status": "passed \| failed",
63	"summary": {
64	"totalSteps": 0,
65	"passed": 0,
66	"failed": 0,
67	"blocked": 0,
68	"notVerifiable": 0,
69	"criticals": 0,
70	"majors": 0,
71	"minors": 0
72	},
73	"agentVerification": [
74	{
75	"step": "Send /start to bot",
76	"tool": "telegram_mcp",
77	"status": "passed \| failed \| not_verifiable",
78	"details": "Bot responded with welcome message"
79	}
80	],
81	"acceptanceCriteria": [
82	{
83	"id": "US-5",
84	"criterion": "Titles generated with correct declensions",
85	"source": "user-spec \| tech-spec \| deferred-from-pre-deploy",
86	"status": "passed \| failed \| blocked",
87	"evidence": "Checked live output, generated title uses correct declension/grammar",
88	"manualVerificationPlan": "Only if blocked — what the user should check, when, how"
89	}
90	],
91	"findings": [
92	{
93	"severity": "critical \| major \| minor",
94	"title": "Bot does not respond to /start",
95	"expected": "Welcome message within 3 seconds",
96	"actual": "No response after 10 seconds",
97	"reproduction": "Steps to reproduce..."
98	}
99	]
100	}
101	```
102
103	Status decision: `passed` if zero criticals, `failed` if one or more criticals.
104
105	## Guidelines
106
107	- Every finding includes concrete reproduction: steps, expected vs actual, tool output.
108	- If all MCP tools are unavailable — report that automated verification is not possible, suggest manual steps.
109	- Empty findings array = clean verification.
110
111	## Final Check
112
113	Before finishing, verify:
114	- [ ] All AVP steps executed or marked `not_verifiable`
115	- [ ] All deferred criteria from pre-deploy QA report addressed
116	- [ ] All acceptance criteria from user-spec and tech-spec verified on live environment
117	- [ ] Every `blocked` criterion has a manual verification plan (what to check, when, how)
118	- [ ] Output JSON is vali

People who liked this also viewed

6 picks

Type