Fullstack Regression Worker

What it does

Example prompts

• “/fullstack-regression-worker”

Workflow steps

10 steps, taken from the first numbered list in SKILL.md.

1. 1Read mission.md, mission AGENTS.md, .factory/services.yaml, and the relevant validation assertions.
2. 2Reproduce the regression first end-to-end so the contradiction is explicit.
3. 3Write the narrowest failing tests needed on both sides
4. 4Keep local runtime on 8080 and 5173, and keep backend assertions bound to the dedicated test schema.
5. 5Use the same seeded/session state across backend and browser proof; do not validate two unrelated scenarios and call them “fullstack.”
6. 6Prefer one reproducible story path
7. 7Run both the backend and frontend commands touched by the feature.
8. 8Use agent-browser for one manual replay of the regression path after automated tests pass. If that replay is not possible because of a…
9. 9Do not claim the procedure was fully followed unless the handoff includes either a real interactiveChecks replay entry or an explicit…
10. 10In the handoff, explain the original contradiction, the shared seeded/session setup, and how the final proof spans both layers.

What it can do on your machine

Read from SKILL.md and the folder at commit 5ffd0a0. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are json).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Fullstack Regression Worker loads about 1.1k tokens when it runs. Until then it costs about 30 tokens; SKILL.md has 349 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~30

When it runs · the whole SKILL.md, loaded when a task matches

~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from CorrectRoadH/OpenTickly at commit 5ffd0a0, republished under its AGPL-3.0 licence (© CorrectRoadH). 349 words, ~1,070 tokens.

name

fullstack-regression-worker

description

Build cross-layer regression coverage that ties browser-visible behavior to backend truth.

{
  "salientSummary": "Closed a stale-current-vs-history regression by pairing one backend truth test with one browser-visible replay. The same seeded timer flow now shows matching current-timer and history state after start, stop, and edit.",
  "whatWasImplemented": "Added a cross-layer regression lane for tracking state consistency. The new coverage uses one shared seeded/session scenario to prove backend current-timer reads and browser-visible timer history no longer contradict each other after the previously flaky flow.",
  "whatWasLeftUndone": "",
  "verification": {
    "commandsRun": [
      {
        "command": "go test ./apps/backend/internal/tracking/... -count=1 -parallel 4",
        "exitCode": 0,
        "observation": "Backend truth tests passed for the regression scenario."
      },
      {
        "command": "vp run test:e2e:website -- e2e/timer-page.spec.ts",
        "exitCode": 0,
        "observation": "Browser regression replay passed on the reused local runtime."
      }
    ],
    "interactiveChecks": [
      {
        "action": "Replayed the start-stop-edit flow in agent-browser and compared the visible history with backend readback.",
        "observed": "The running header cleared at the same time history showed the stopped entry, with no stale contradiction."
      }
    ]
  },
  "tests": {
    "added": [
      {
        "file": "apps/website/e2e/timer-page.spec.ts + apps/backend/internal/tracking/...",
        "cases": [
          {
            "name": "current timer and history stay aligned after stop and edit",
            "verifies": "VAL-REG-004"
          }
        ]
      }
    ]
  },
  "discoveredIssues": []
}

© CorrectRoadH, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .factory/skills/fullstack-regression-worker of CorrectRoadH/OpenTickly.

Open the folder on GitHubat commit 5ffd0a0