DRAFT - [CACHE FOLDER] Fix cache folder using rules in skills

skills/automation-coverage/SKILL.md

@@ -30,7 +30,11 @@ This skill works **only with automated e2e tests** (Playwright, Cypress, Webdriv
 - **DO NOT** process unit tests.
 - **DO NOT** process manual markdown test cases (use `manual-coverage` instead - if already exists).
 - **DO NOT** suggest creating new tests.
-- **Only touch two files in this repo.** It may write `coverage.e2e.yml` (or the path the user gave) and add one `.testclaw-context/` line to `.gitignore` if it is missing. Nothing else — never a source or test file. If the e2e tests live in another repo, clone it into the gitignored `.testclaw-context/e2e-tests/`, never into a tracked folder (see Step 1).
+- **Only touch two files in this repo.** It may write `coverage.e2e.yml` (or the path the user gave) and add one `.testclaw-context/` line to `.gitignore` if it is missing. Nothing else — never a source or test file. If the e2e tests live in another repo, clone it into:
+  - **Big project with source code** → gitignored `.testclaw-context/e2e-tests/`
+  - **Test-only project** (e2e repo without `src/`) → tracked folder (e.g., `e2e-tests/`)
+
+Never clone into a tracked folder in a source-code repo.
 - **Don't write scripts. Never use Python.** Read test files with your file tool; pull out IDs and tags with `grep` (Step 3). To check the finished coverage file, pipe it through `js-yaml` into the one tiny bundled helper (symlinked from `manual-coverage`): `npx js-yaml coverage.e2e.yml | node scripts/check-coverage.mjs` (Step 6). That's the only script. If you ever need more than a `grep`, a one-line `node -e '…'` is the limit — never `python`, never a parser of your own.
 
 ---

skills/manual-coverage/SKILL.md

@@ -29,7 +29,12 @@ This skill works **only with manual tests in markdown format**.
 
 - **DO NOT** process unit, functional, or e2e test files.
 - **DO NOT** suggest creating new automated tests.
-- **Only touch two files in this repo.** This skill runs inside the user's source-code repo. It may write `coverage.manual.yml` (or the path the user gave) and add one `.testclaw-context/` line to `.gitignore` if it is missing. Nothing else — never a source file. Cases pulled from Testomat.io go into the gitignored `.testclaw-context/manual-tests/`, never into a tracked folder (see Step 1).
+- **Only touch two files in this repo.** This skill runs inside the user's source-code repo. It may write `coverage.manual.yml` (or the path the user gave) and add one `.testclaw-context/` line to `.gitignore` if it is missing. Nothing else — never a source file. Cases pulled from Testomat.io go into:
+  - **Big project with source code** → gitignored `.testclaw-context/manual-tests/`
+  - **Test-only project** (no `src/`) → tracked `manual-tests/`
+
+Never pull cases into a tracked folder in a source-code repo.
+
 - **Don't write scripts. Never use Python.** Read `.test.md` files with your file tool; pull out IDs and tags with `grep` (Step 2). To check the finished coverage file, pipe it through `js-yaml` into the one tiny bundled helper: `npx js-yaml coverage.manual.yml | node scripts/check-coverage.mjs` (Step 5). If more checks needed use custom script with `npx js-yaml`.
 
 If automated test files (e.g. e2e test, unit, api)  are encountered while exploring, ignore them and continue with the manual markdown set.
@@ -49,11 +54,14 @@ From the `project-scan` result, capture:
 
 If `project-scan` reports **no manual tests** (it checks the cache too, so this means nothing was pulled before):
 - ❓ Ask the user how to proceed:
-  1. Pull cases from Testomat.io — have **`sync-cases`** pull into the gitignored cache: `npx check-tests pull -d .testclaw-context/manual-tests`, then add `.testclaw-context/` to `.gitignore` if missing. Re-run `project-scan` and continue.
+  1. Pull cases from Testomat.io — have **`sync-cases`** pull into the appropriate location:
+     - **Source-code project** (has `src/`) → gitignored `.testclaw-context/manual-tests/`
+     - **Test-only project** (no `src/`) → tracked `manual-tests/`
+     Then add `.testclaw-context/` to `.gitignore` if needed. Re-run `project-scan` and continue.
   2. Point to a directory the scan missed (then re-run `project-scan` there).
   3. Stop.
 
-Never pull cases into a tracked folder. Don't repeat `sync-cases` pull logic here.
+Never pull cases into a tracked folder in a source-code repo. Don't repeat `sync-cases` pull logic here.
 
 ### Step 2: Extract test information
 
@@ -207,7 +215,7 @@ It's the only script — everything else is `grep`, your file tool, or `npx js-y
 
 ### Recovery
 
-- **No manual tests found (cache included)** → have `sync-cases` pull into `.testclaw-context/manual-tests/`, or ask the user for a directory the scan missed.
+- **No manual tests found (cache included)** → have `sync-cases` pull (auto-detects location based on project type), or ask the user for a directory the scan missed.
 - **Markdown files with no `@S`/`@T` IDs** → ask whether to push first via `sync-cases`, or skip those files.
 - **Ambiguous source layout** → ask the user which directories are application code.
 
@@ -225,7 +233,7 @@ It's the only script — everything else is `grep`, your file tool, or `npx js-y
 ```
 Use manual-coverage skill to build coverage.manual.yml for our manual cases
 ```
-If there are no local `.test.md` files, it pulls them into the gitignored `.testclaw-context/manual-tests/` and works from there.
+If there are no local `.test.md` files, it pulls them into the appropriate location (auto-detected: `.testclaw-context/manual-tests/` for source-code projects, `manual-tests/` for test-only projects) and works from there.
 
 **Cases already local:**
 ```
@@ -238,7 +246,7 @@ Use manual-coverage skill, output to ops/coverage.qa.yml
 ```
 
 **Full workflow (source repo):**
-1. `manual-coverage` runs `project-scan`. No local cases, so it has `sync-cases` pull into `.testclaw-context/manual-tests/` (gitignored) and re-runs `project-scan`.
+1. `manual-coverage` runs `project-scan`. No local cases, so it has `sync-cases` pull (auto-detects location) and re-runs `project-scan`.
 2. It maps source files to suite/test/tag IDs and writes `coverage.manual.yml`. The only tracked changes are that file and one line in `.gitignore`.
 3. `npx @testomatio/reporter run --kind manual --filter "coverage:file=coverage.manual.yml,diff=main"` creates a pending run with only the affected cases.
 
@@ -248,6 +256,7 @@ Use manual-coverage skill, output to ops/coverage.qa.yml
 
 | Action                                 | Command                                                                                              |
 | -------------------------------------- | ---------------------------------------------------------------------------------------------------- |
-| Pull cases into the gitignored cache  | `npx check-tests pull -d .testclaw-context/manual-tests`                                            |
+| Pull cases into cache (source-code project) | `npx check-tests pull -d .testclaw-context/manual-tests` (auto-detected) |
+| Pull cases into tracked folder (test-only project) | `npx check-tests pull -d manual-tests` (auto-detected) |
 | Create affected manual run             | `npx @testomatio/reporter run --kind manual --filter "coverage:file=coverage.manual.yml,diff=main"`  |
 | Group runs                             | `TESTOMATIO_RUNGROUP="Regression 911" npx @testomatio/reporter run --kind manual --filter "..."`     |

skills/project-scan/SKILL.md

@@ -56,25 +56,24 @@ Where is the source code?
 
 - If user provides `Git repo` or `another folder` variant => clone or symlink it into `.testclaw-context/code/` (see the rule below). No-op if it's already there.
 
-#### Rule for pulled data — `.testclaw-context/`
+#### Rule for pulled data — test directory detection
 
-**This applies to every skill, not just `project-scan`.**
+**When you need to pull external data** (manual test cases, app code, e2e tests from another repo) into this project:
 
-1. First, see what's already in the current folder.
-2. If you need data that isn't here — manual test cases, the application code, e2e tests, a repo to clone — pull it into `.testclaw-context/` (gitignored). **Never** pull it into a regular folder like `manual-tests/` or `automated-tests/`; that pollutes the repo.
-3. If the data is already in the repo, use it where it is — you don't need `.testclaw-context/` for that.
+| This project has… | Store pulled data in… | Gitignored? |
+| ------------------ | ---------------------- | ----------- |
+| **Source code** (`src/` folder) or **big project** | `.testclaw-context/…` | **Yes** |
+| **Only test infrastructure** (e2e dirs like `tests/`, `playwright/`, `cypress/` but no `src/`) | `manual-tests/` or `e2e-tests/` | No |
+| **Empty / manual-only** | `manual-tests/` | No |
 
-| When you run inside…    | What you'd pull in…  | Goes in…                          |
-| ----------------------- | -------------------- | --------------------------------- |
-| a source-code repo      | manual test cases    | `.testclaw-context/manual-tests/` |
-| a manual-tests repo     | the application code | `.testclaw-context/code/`         |
-| (e2e tests in own repo) | the e2e tests        | `.testclaw-context/e2e-tests/`    |
+**Detection logic:**
+1. Has `src/` folder or is a monorepo? → Use `.testclaw-context/`
+2. Has e2e test dirs (`tests/`, `playwright/`, `cypress/`, `e2e/`)? → Use tracked folder (`manual-tests/`, `e2e-tests/`)
+3. Otherwise → Default to `manual-tests/`
 
-Any skill that creates `.testclaw-context/...` must also add `.testclaw-context/` to the project's `.gitignore` — but only if it is not there yet. Never add it twice.
+> **Override:** If user specifies a path, use that instead.
 
-(The rule is about *pulled* data. Output a skill *produces* and the user wants — a `coverage.*.yml`, new `*.test.md` files — goes in the repo as normal.)
-
-On a source repo with no manual tests, `project-scan` changes no tracked file: it only creates the gitignored `.testclaw-context/` directory and, if needed, adds one line to `.gitignore`.
+**Only touch `.testclaw-context/` and tracked folders — never pollute repo with cache in wrong location.**
 
 ---
 

skills/sync-cases/SKILL.md

@@ -1,14 +1,14 @@
 ---
 name: sync-cases
-description: Synchronize test scenarios and cases between a local project and Testomat.io. Use this skill whenever the user wants to pull/export/download tests from Testomat.io; or push/import/sync new or updated test cases back to the TMS in corresponding `*.test.md` format. Supports custom directories, markdown test format and advanced import/export workflows.
+description: Synchronize test scenarios and cases between a local project and Testomat.io. Use this skill whenever the user wants to pull/export/download tests from Testomat.io; or push/import/sync new or updated test cases back to the TMS in corresponding `*.test.md` format. Supports custom directories, markdown test format and advanced import/export workflows. Automatically detects project type and stores test cases in `manual-tests/` (for simple/test projects) or `.testclaw-context/manual-tests/` (for projects with source code).
 inputs:
   testDir:
-    description: "Where to put pulled tests (default: `.testclaw-context/manual-tests` — the gitignored cache). Only override with a tracked path if the repo already keeps its `*.test.md` files there."
+    description: "Override auto-detected test directory. By default: `manual-tests/` for simple projects, `.testclaw-context/manual-tests/` for big projects with additional backend, frontend logic"
     required: false
 license: MIT
 metadata:
   author: Testomat.io
-  version: 1.0.0
+  version: 1.1.0
 ---
 
 # SYNC-CASES SKILL: What I do
@@ -57,7 +57,29 @@ Ask the user to obtain it from Testomat.io project:
 - Navigate to **Settings → Project → Project Reporting API key**
 _( Project path example by "project-id": `https://app.testomat.io/projects/<project-id>/settings/project` )_
 
-### Step 2: Pull or Push Operations
+### Step 2: Detect Project Type and Choose Test Directory
+
+Before pulling or pushing, detect the project type to determine the correct test case location.
+
+#### Test Directory Rules
+
+| Project Type | Test Cases Location | Gitignored? |
+| ------------ | ------------------- | ----------- |
+| **Manual-only** (empty project, no backend, frontend logic) OR **E2E/test project** (`tests/`, `playwright/`, `cypress/`, `e2e/`) | `manual-tests/` | No |
+| **Has additional source code** (has backend, frontend logic) or **big project** (monorepo, multiple test types) | `.testclaw-context/manual-tests/` | **Yes** — add to `.gitignore` if missing |
+| **Folder from the user's prompt/request** | `testDir` | No |
+
+#### Detection Logic
+
+1. Has backend, frontend, extra logic code? => Use `.testclaw-context/manual-tests/`
+2. Has `tests/`, `playwright/`, `cypress/`, `e2e/`, `tests-e2e/`? => Use `manual-tests/`
+3. Otherwise => Default to `manual-tests/`
+
+> **Override:** If `testDir` input is provided or user specifies `-d <path>` in command, use that path instead.
+
+---
+
+### Step 3: Pull or Push Operations
 
 #### Always use latest version of check-tests
 
@@ -76,10 +98,12 @@ Download/Retrieves test scenarios from Testomat.io and saves them as Markdown fi
 - Refactor test cases offline.
 - Export only some specific suites by id.
 
-**Where to pull:** into the gitignored cache `.testclaw-context/manual-tests/`, and add `.testclaw-context/` to the project `.gitignore` if it is not there yet. Pulled cases must never land in a tracked folder (`manual-tests/`, etc.) — that pollutes the repo. You can still edit them there and push back; being gitignored doesn't stop that. (If the repo *already* keeps its `*.test.md` files in a tracked folder, you don't need to pull at all — work with them where they are, or pass `-d <that folder>` for an in-place refresh.) This matches `project-scan`, which pulls the *code* into `.testclaw-context/code/` when it runs inside a manual-tests repo.
+**Where to pull:** Determined by [Step 2: Detect Project Type](#step-2-detect-project-type-and-choose-test-directory). The target directory is created if it doesn't exist.
 
 **Pre-Pull:**
-- Create `.testclaw-context/manual-tests/` if it doesn't exist; ensure `.testclaw-context/` is gitignored.
+1. Run detection (if `testDir` not provided)
+2. Create target directory if it doesn't exist
+3. If target is `.testclaw-context/` → ensure it's gitignored
 
 **Command:**
 ```bash
@@ -97,13 +121,16 @@ Optional variant - **pull by specific suite-ids**
 
 **Pull Examples:**
 ```bash
-# Default — pull into the gitignored cache
-npx check-tests pull -d .testclaw-context/manual-tests
+# Auto-detected location
+npx check-tests pull
 
-# Repo that already keeps its test cases tracked — refresh them in place
+# Manual-only or E2E project → manual-tests/
 npx check-tests pull -d manual-tests
 
-# Pull specific suites by IDs
+# Big project with extra source code → gitignored cache
+npx check-tests pull -d .testclaw-context/manual-tests
+
+# Specific suite IDs
 npx check-tests pull --suite-ids "@S12345678,@S87654321"
 ```
 
@@ -156,16 +183,22 @@ npx check-tests push [-d <directory>] [--files <files...>]
 
 **IMPORTANT:** When the files to push are known (e.g. just produced by `generate-cases` / `improve-test-cases`), pass them explicitly via `--files` (alias `-f`). Without `--files` the CLI falls back to the default glob `**/*.test.md`, which may pick up unrelated files. Quote glob patterns. Paths resolve relative to `--dir`.
 
-**Examples:**
+**Push Examples:**
 ```bash
-# Specific files (preferred when known — e.g. just produced by generate-cases)
+# Push from manual-tests/
+npx check-tests push -d manual-tests
+
+# Push from gitignored cache
+npx check-tests push -d .testclaw-context/manual-tests
+
+# Specific files from cache folder (preferred when known — e.g. just produced by generate-cases)
 npx check-tests push --files .testclaw-context/manual-tests/login.test.md .testclaw-context/manual-tests/checkout.test.md
 
 # Custom glob
 npx check-tests push --files ".testclaw-context/manual-tests/**/*.test.md"
 
-# Default glob (**/*.test.md) under -d — cases you pulled and edited in the cache
-npx check-tests push -d .testclaw-context/manual-tests
+# Specific files not from cache .testclaw-context folder
+npx check-tests push --files login.test.md checkout.test.md
 ```
 
 **Important constraints:**
@@ -177,8 +210,6 @@ npx check-tests push -d .testclaw-context/manual-tests
 
 #### Labels Handling (Intent-Based)
 
-Use `TESTOMATIO_LABELS` in sync/push **only if the user explicitly requests to set or override labels** in their query.
-
 **Triggers:**
 * "push tests with labels smoke".
 * "import tests to TMS and set label=regression":
@@ -193,7 +224,7 @@ After completing sync operations, output a short log-style summary:
 ```
 Sync Complete:
 - Action: pull/push
-- Directory: .testclaw-context/manual-tests
+- Directory: <auto-detected path>
 - Tests synced: 15
 - Status: Success
 ```
@@ -231,7 +262,7 @@ Stop execution if:
 
 ## Examples
 
-**Pull tests** (lands in the gitignored `.testclaw-context/manual-tests/`):
+**Pull tests:**
 ```
 Use sync-cases skill to pull tests from Testomat.io
 ```
@@ -252,7 +283,8 @@ Use sync-cases to push tests to Testomat.io
 
 | Action          | Command                                                              |
 | --------------- | -------------------------------------------------------------------- |
-| Pull            | `npx check-tests pull -d manual-tests` (the default; gitignored) |
+| Pull            | `npx check-tests pull -d <dir>`                                      |
+| Push            | `npx check-tests push -d <dir>`                                      |
 | Push (files)    | `npx check-tests push --files <file1.test.md> <file2.test.md>`       |
 | Push (glob)     | `npx check-tests push --files "<dir>/**/*.test.md"`                  |
 | Push (default)  | `npx check-tests push -d <directory>` (glob: `**/*.test.md`)         |