QA workflow for testing a feature or task implementation. Analyzes changes, produces test plans, and describes concrete test scenarios. Use when user says "test this", "write test plan", "what should I test", or "QA this branch".
80%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Passed
No findings from the security scan
Fix and improve this skill with Tessl
tessl review fix ./skills/aif-qa/SKILL.mdApply this contract whenever the workflow asks the user to decide:
AskUserQuestion and Questions below describe the intended interaction, not a required tool name. Use only a question tool actually exposed and permitted in the current session/mode, with its advertised schema. Do not guess aliases, probe unavailable tools, or assume questionnaire exists. A frontmatter entry alone does not make a tool available.HANDOFF_MODE=1), do not prompt the user directly. Use declared defaults only for optional preferences within the authorized scope and report assumptions. If a required decision, approval, or material scope/behavior choice remains unresolved, stop dependent work and return the missing decision and options to the parent/caller using its existing blocked/failure contract. Never invent consent or select a new scope to avoid blocking.This delivery policy applies to this skill and its linked references. Preserve workflow-specific question scope, required confirmations, approval boundaries, and blocked-result rules.
Before mode routing or work, read existing files relative to the current project root in this order (reuse them if already loaded):
.ai-factory/skill-context/global-skill-context.md — all bundled skills..ai-factory/skill-context/aif-qa/SKILL.md — this skill.Missing files are skipped; an existing unreadable file stops dependent work with its path.
Precedence: individual > build (when applicable) > global > skill defaults; retain non-conflicting instructions and check all outputs against the effective rules.
These layers customize workflow; they do not replace the project's rules/architecture hierarchy or relax explicit requirements, ownership, scope, safety/privacy, validation, or approval boundaries. Report conflicts with those boundaries instead of overriding them.
global is a project-local layer, not home-directory configuration. Custom/extension skills do not automatically use it.
Generates change summaries, produces test plans, and describes test scenarios for a feature or task implementation.
The skill operates in three sequential modes.
--all skips inter-stage prompts only. Missing-context questions and required
approval decisions still follow Runtime-aware questions; the flag does not
make the session background/non-interactive or supply consent.
| Argument | Mode | What you do |
|---|---|---|
change-summary | Change summary | Analyze what changed, assess risks, produce a summary |
test-plan | Test plan | Create a structured test plan based on the change summary |
test-cases | Test cases | Describe concrete test scenarios based on the plan |
--all | Full pipeline | Run all three modes in sequence without prompting between stages |
FIRST: Read .ai-factory/config.yaml if it exists to resolve:
paths.description, paths.architecture, paths.qa (default: .ai-factory/qa)language.ui for AskUserQuestion prompts, progress messages, final summaries, and next-step guidancelanguage.artifacts for generated QA artifactslanguage.technical_terms for human-readable technical terminology style when generating artifactslanguage.artifacts is missing, use language.uiengit.enabled and git.base_branch for branch comparisonIf config.yaml doesn't exist, use defaults:
.ai-factory/DESCRIPTION.md.ai-factory/ARCHITECTURE.md.ai-factory/qa/ui_language: enartifact_language: entechnical_terms_policy: keeptruemainStore:
ui_language = language.ui || "en"artifact_language = language.artifacts || language.ui || "en"technical_terms_policy = language.technical_terms || "keep"git_enabled = git.enabled when present, otherwise truebase_branch = git.base_branch || "main"All AskUserQuestion prompts, user-visible explanations, stage completion messages, and next-step guidance MUST be written in ui_language.
All generated artifacts (change-summary.md, test-plan.md, test-cases.md) MUST be written in artifact_language.
Templates define structure, not language. Use the canonical English templates in templates/*.md. If artifact_language is not en, translate them to artifact_language before saving: headings, labels, checklist items, placeholders, enum labels, risk labels, and explanatory text must be in artifact_language.
Do not use English templates verbatim when artifact_language is not en. Preserve markdown structure, table shapes, checkbox syntax, test case IDs (TC-001), code identifiers, paths, commands, branch names, config keys, API names, package names, and raw error messages.
For artifact_language = ru, write human-readable prose, headings, risks, priorities, recommendations, test steps, and expected results in Russian. Keep code identifiers, filenames, branch names, commands, config keys, API names, and raw error text unchanged.
Apply technical_terms_policy while writing artifacts:
keep — keep common technical terms such as commit, branch, diff, endpoint, payload, rollback, regression, and fixture when that is clearer for the project audiencetranslate — translate human-readable technical terms where a natural target-language term existsmixed — translate ordinary prose terms while keeping code, infrastructure, and ecosystem terms unchangedIf git_enabled = false or the current directory is not a git work tree, do not run git diff/log commands. Use manual change context mode instead: ask the user in ui_language to provide one of these sources of change context before running change-summary: pasted diff, changed file list, short implementation description, or cancel.
Read the resolved description path if the file exists, to understand:
Read the resolved architecture path if the file exists, to understand:
Use this context when generating summaries, test plans, and test cases.
Read .ai-factory/skill-context/aif-qa/SKILL.md — MANDATORY if the file exists.
This file contains project-specific rules accumulated by /aif-evolve from patches,
codebase conventions, and tech-stack analysis. These rules are tailored to the current project.
How to apply skill-context rules:
Parse $ARGUMENTS fully before doing anything else:
--all flag — if present, set all_mode = true and remove the flag from argumentschange-summary, test-plan, or test-cases; remove it from argumentsResolve the working branch:
If git_enabled = false or the repository is not a git work tree:
If branch was provided in arguments → use it as the resolved branch label
Otherwise → set resolved_branch = "manual"
Use manual change context mode for analysis
If git_enabled = true and the repository is a git work tree:
If branch was provided in arguments → use it as the resolved branch
Otherwise → run: git branch --show-currentStore both values for use in all reference files:
resolved_branch — the branch being analyzed (used to locate/save artifacts)
artifact_dir — <resolved paths.qa>/<branch-slug>, where branch-slug is a deterministic, filesystem-safe, collision-resistant slug derived from resolved_branch. Compute it in three steps:
resolved_branch and replace every character that is not in [A-Za-z0-9._-] with -, collapse runs of consecutive - into a single -, and trim leading/trailing -. If the result is empty, use branch. Then MUST truncate to the first 40 ASCII characters. Because the normalized safe_slug alphabet is [A-Za-z0-9._-], byte length and character length are identical. Call this safe_slug.git hash-object --stdin <<< "<resolved_branch>" and take the first 8 hex characters of the output. Call this hash8. The hash is derived from the original, unnormalized branch name so branches that collapse to the same safe_slug still produce different derived slugs in normal use.branch-slug = "<safe_slug>-<hash8>".Why the hash: a readable slug alone is lossy — feature/foo and feature-foo normalize to the same safe_slug and would overwrite each other's artifacts. Appending a short hash of the full original name keeps the derived slug stable, readable, and collision-resistant for practical branch naming.
Examples:
feature/foo → safe_slug=feature-foo, hash8=a72ccce7 → feature-foo-a72ccce7feature-foo → safe_slug=feature-foo, hash8=6f80dfc6 → feature-foo-6f80dfc6main → safe_slug=main, hash8=<computed> → main-<hash8>all_mode — whether to skip inter-stage prompts
If no mode was provided and all_mode = false — ask the user in ui_language:
AskUserQuestion in `ui_language`.
Meaning: ask which QA mode to run.
Options meaning:
1. Change summary (`change-summary`) — analyze what changed, assess risks, produce a summary
2. Test plan (`test-plan`) — create a structured test plan based on the change summary
3. Test cases (`test-cases`) — describe concrete test scenarios based on the plan
4. Full pipeline (`--all`) — run all three modes in sequenceThe skill runs strictly sequentially — each stage uses the artifact from the previous one:
change-summary → test-plan → test-casesRead the detailed instructions for the selected mode:
Read references/CHANGE-SUMMARY.md
Read references/TEST-PLAN.md
Read references/TEST-CASES.md
Run all three modes in sequence. After each stage completes successfully,
proceed to the next automatically — do NOT show the inter-stage AskUserQuestion.
1. Execute change-summary (references/CHANGE-SUMMARY.md) → save artifact
2. Execute test-plan (references/TEST-PLAN.md) → save artifact
3. Execute test-cases (references/TEST-CASES.md) → save artifact
4. Show context cleanup prompt (Step 6 of TEST-CASES.md)If any stage fails (e.g. git error, diff too large and user cancels) — stop the pipeline and report which stage failed.
Do not replace the manual QA plan with automated test implementation details. You may mention existing automated checks only as supporting verification; the primary output must remain manual QA scenarios.
| Priority | When to use |
|---|---|
| High | Core business logic, user data, payments, security, authorization |
| Medium | Supporting functionality, UI/UX, reports, integrations |
| Low | Cosmetic changes, rare scenarios, nice-to-have |
<paths.qa>/<branch-slug>/ — specifically change-summary.md, test-plan.md, and test-cases.md. The --all flag respects the same boundary.paths.description, paths.architecture, paths.qa, language.ui, language.artifacts, language.technical_terms, git.enabled, and git.base_branch; never writes config.yaml.test-plan without a change-summary artifacttest-cases without a test-plan artifactlee-to/ai-factory
eabdfe9
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.