agentleFS
haruharu/redmine_ai_helper/.claude/skills/redmine-playwright-login/SKILL.md

redmine-playwright-login

Skill107 starsChanged 9 months agoMITView on GitHub

Connect to the running Redmine instance and log in via the Playwright MCP browser, using REDMINE_PLAYRIGHT_URL/REDMINE_PLAYRIGHT_USER/REDMINE_PLAYRIGHT_PASSWORD environment variables. Use whenever the user asks to open, check, access, or test something in the running Redmine UI (e.g. verifying a feature in the browser, taking a screenshot of a page, manually exercising a flow).

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 2 are distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If a path is given, open it after logging in.

    line 19
  2. Very specific:

    Read these three environment variables (e.g.

    line 25
  3. Distinctive:

    If any of them is unset or empty, stop and tell the user which one is missing instead of guessing a value.

    line 31
  4. Very specific:

    Use mcp__playwright__browser_navigate to open <REDMINE_PLAYRIGHT_URL>/login.

    line 35
  5. Very specific:

    Use mcp__playwright__browser_snapshot to get element refs for the Login and Password textboxes and the Login button.

    line 39
  6. Very specific:

    Use mcp__playwright__browser_fill_form to fill:

    line 43
  7. Very specific:

    If the user (or $ARGUMENTS) specified a page to open next, navigate there now with mcp__playwright__browser_navigate.

    line 55
  8. Distinctive:

    Do not print the password anywhere except as needed to pass it into the Playwright fill call.

    line 62
  9. Very specific:

    Never save screenshots (mcp__playwright__browser_take_screenshot) or snapshot files into the plugin's working directory — it's a git repo, …

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Redmine Playwright Login

Log in to the currently running Redmine instance through the Playwright MCP browser tools, so the browser session can then be used to inspect pages, verify features, or take screenshots.

User Input

$ARGUMENTS

If a path is given, open it after logging in. Otherwise stop at the post-login page (/my/page).

Procedure

Step 1: Read the Connection Env Vars

Read these three environment variables (e.g. via printenv):

  • REDMINE_PLAYRIGHT_URL — base URL of the running Redmine instance (http://localhost:3000 in the devcontainer)
  • REDMINE_PLAYRIGHT_USER — login name
  • REDMINE_PLAYRIGHT_PASSWORD — password

If any of them is unset or empty, stop and tell the user which one is missing instead of guessing a value.

Step 2: Navigate to the Login Page

Use mcp__playwright__browser_navigate to open <REDMINE_PLAYRIGHT_URL>/login.

Step 3: Inspect the Form

Use mcp__playwright__browser_snapshot to get element refs for the Login and Password textboxes and the Login button.

Step 4: Fill and Submit

Use mcp__playwright__browser_fill_form to fill:

  • the Login textbox with REDMINE_PLAYRIGHT_USER
  • the Password textbox with REDMINE_PLAYRIGHT_PASSWORD

Then mcp__playwright__browser_click the Login button.

Step 5: Verify

Confirm the resulting page URL is no longer /login (typically redirects to /my/page). If it's still on /login, the snapshot will show a flash error — report the failure to the user rather than retrying blindly.

Step 6: Optional Navigation

If the user (or $ARGUMENTS) specified a page to open next, navigate there now with mcp__playwright__browser_navigate.

Notes

  • The Playwright MCP server launches a headless Chromium (/usr/bin/chromium) inside the devcontainer, so Redmine is reached at http://localhost:3000. There is no separate browser container (ADR-066).
  • Files passed to mcp__playwright__browser_file_upload are read from the devcontainer's filesystem; use absolute paths there (e.g. under the scratchpad directory).
  • This is a local/test Redmine instance for development — credentials come from env vars set up for this purpose, not real user secrets.
  • Do not print the password anywhere except as needed to pass it into the Playwright fill call.
  • Reuse the already-open browser session for subsequent actions in the same conversation — no need to re-login unless the session was closed or logged out.
  • Never save screenshots (mcp__playwright__browser_take_screenshot) or snapshot files into the plugin's working directory — it's a git repo, and stray files there risk being committed. Always write them under the session scratchpad directory (or another path under /tmp) instead.

Error Handling

  • Missing env var: stop and ask the user to set it, don't fall back to a default.
  • Login failure (still on /login after submit, or an error flash message shown): report the exact error text from the page snapshot to the user.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.