Non-negotiable rules:
- Turn the incoming spec into explicit scenarios before testing.
- Confirm ambiguous targets or platforms before execution.
- Use the
browse skill for actual browser or simulator interaction.
- Capture evidence for each failed or important scenario.
- Save rerunnable flows only when they are cleanly scoped and valuable for regression.
browse-qa
Inputs
$request: Ticket, URL, app target, acceptance criteria, or plain-language QA request
Goal
Produce a credible QA run that:
- turns a vague request into explicit scenarios
- runs those scenarios against the real target
- captures evidence for passes and failures
- saves rerunnable browse flows when appropriate
- returns a clean QA report the user can act on
Step 0: Resolve the target and spec
Work out:
- what feature is being tested
- the platform: web, iOS simulator, Android emulator, or macOS app
- the target URL, build artifact, or app identifier
- the acceptance criteria or expected outcomes
If the target or platform is ambiguous, ask before testing.
Success criteria: The QA target and acceptance criteria are explicit.
Step 1: Build the scenario list
Break the request into:
- happy path
- validation or error path
- edge cases
- exploratory probes worth trying after the stated acceptance criteria
Keep the plan small enough to execute in one session unless the user asked for a broad QA sweep.
Success criteria: There is a concrete scenario list with expected outcomes.
Step 2: Execute with the browse skill
Use Skill to invoke browse for:
- navigation
- page stabilization
- interactive element discovery
- clicks, fills, and submissions
- screenshots, console logs, and network inspection
Rules:
- keep the same browse session for one logical user journey unless isolation is intentional
- use headed or handoff mode only for real blockers like CAPTCHA, MFA, or OAuth walls
- save screenshots and flow artifacts under the browse session directories
Success criteria: Each scenario is executed against the real target with enough evidence to judge pass or fail.
Step 3: Record reusable regression flows
When the scenario is stable and worth rerunning:
- record the flow
- split unrelated user journeys into separate flow files
- give each saved flow a descriptive name
Do not save noisy or half-manual recordings as regression artifacts.
Success criteria: Saved flows are independently rerunnable and map to clear scenarios.
Step 4: Report results
For each scenario, record:
- expected result
- actual result
- pass or fail
- evidence such as screenshot path, console error, or network symptom
Also report:
- exploratory findings
- regression flow files created
- blockers that prevented full verification
Success criteria: Another engineer can understand the QA outcome without replaying the whole session first.
Guardrails
- Do not duplicate the full browse command manual inline; use the
browse skill.
- Do not guess a platform when the request is ambiguous.
- Do not save low-signal flows just to say a regression asset was created.
- Do not add
disable-model-invocation; QA should stay available when the user asks for it.
- Do not add
context: fork; results are usually needed in the current execution flow.
Output Contract
Report:
- target tested
- scenarios executed
- pass/fail result for each scenario
- evidence captured
- regression flow files created
- blockers or uncovered gaps
1---
2name: browse-qa
3version: 2.0.0
4description: |
5 Turn a ticket, URL, or plain-language spec into an executed QA run AND reusable regression flows —
6 scenarios and evidence, not ad-hoc clicking — by decomposing the request into happy/error/edge/
7 exploratory scenarios, driving each against the real target through the `browse` skill (web or
8 iOS/Android/macOS), capturing pass/fail evidence, and recording cleanly-scoped browse flows worth
9 rerunning. Saves a flow only when it is well-scoped and valuable, never noisy half-recordings. Use
10 to QA a feature and leave behind rerunnable coverage.
11allowed-tools:
12 - AskUserQuestion
13 - Bash
14 - Read
15 - Write
16 - Skill
17argument-hint: "[feature, ticket, URL, or QA target]"
18arguments:
19 - request
20when_to_use: |
21 Use when the user asks to QA a feature, validate a story, test a page or app flow, or convert manual
22 QA into rerunnable coverage. Examples: "QA this checkout flow", "validate LINEAR-123", "turn this
23 manual QA into browse flows". Do NOT use for pure static code review or bug-hunting a diff (find-bugs
24 / branch-review-before-pr), nor as the low-level browser driver itself (browse); this owns the QA
25 plan, evidence, and regression artifacts, and delegates the clicking to browse.
26effort: high
27metadata:
28 source: https://skillmd.com/skills/ulpi-io/browse-qa
29---
30
31<EXTREMELY-IMPORTANT>
32This skill owns the QA plan and report, not the low-level browser command encyclopedia.
33
34Non-negotiable rules:
351. Turn the incoming spec into explicit scenarios before testing.
362. Confirm ambiguous targets or platforms before execution.
373. Use the `browse` skill for actual browser or simulator interaction.
384. Capture evidence for each failed or important scenario.
395. Save rerunnable flows only when they are cleanly scoped and valuable for regression.
40</EXTREMELY-IMPORTANT>
41
42# browse-qa
43
44## Inputs
45
46- `$request`: Ticket, URL, app target, acceptance criteria, or plain-language QA request
47
48## Goal
49
50Produce a credible QA run that:
51
52- turns a vague request into explicit scenarios
53- runs those scenarios against the real target
54- captures evidence for passes and failures
55- saves rerunnable browse flows when appropriate
56- returns a clean QA report the user can act on
57
58## Step 0: Resolve the target and spec
59
60Work out:
61
62- what feature is being tested
63- the platform: web, iOS simulator, Android emulator, or macOS app
64- the target URL, build artifact, or app identifier
65- the acceptance criteria or expected outcomes
66
67If the target or platform is ambiguous, ask before testing.
68
69**Success criteria**: The QA target and acceptance criteria are explicit.
70
71## Step 1: Build the scenario list
72
73Break the request into:
74
75- happy path
76- validation or error path
77- edge cases
78- exploratory probes worth trying after the stated acceptance criteria
79
80Keep the plan small enough to execute in one session unless the user asked for a broad QA sweep.
81
82**Success criteria**: There is a concrete scenario list with expected outcomes.
83
84## Step 2: Execute with the `browse` skill
85
86Use `Skill` to invoke `browse` for:
87
88- navigation
89- page stabilization
90- interactive element discovery
91- clicks, fills, and submissions
92- screenshots, console logs, and network inspection
93
94Rules:
95
96- keep the same browse session for one logical user journey unless isolation is intentional
97- use headed or handoff mode only for real blockers like CAPTCHA, MFA, or OAuth walls
98- save screenshots and flow artifacts under the browse session directories
99
100**Success criteria**: Each scenario is executed against the real target with enough evidence to judge pass or fail.
101
102## Step 3: Record reusable regression flows
103
104When the scenario is stable and worth rerunning:
105
106- record the flow
107- split unrelated user journeys into separate flow files
108- give each saved flow a descriptive name
109
110Do not save noisy or half-manual recordings as regression artifacts.
111
112**Success criteria**: Saved flows are independently rerunnable and map to clear scenarios.
113
114## Step 4: Report results
115
116For each scenario, record:
117
118- expected result
119- actual result
120- pass or fail
121- evidence such as screenshot path, console error, or network symptom
122
123Also report:
124
125- exploratory findings
126- regression flow files created
127- blockers that prevented full verification
128
129**Success criteria**: Another engineer can understand the QA outcome without replaying the whole session first.
130
131## Guardrails
132
133- Do not duplicate the full browse command manual inline; use the `browse` skill.
134- Do not guess a platform when the request is ambiguous.
135- Do not save low-signal flows just to say a regression asset was created.
136- Do not add `disable-model-invocation`; QA should stay available when the user asks for it.
137- Do not add `context: fork`; results are usually needed in the current execution flow.
138
139## Output Contract
140
141Report:
142
1431. target tested
1442. scenarios executed
1453. pass/fail result for each scenario
1464. evidence captured
1475. regression flow files created
1486. blockers or uncovered gaps