sztadii/agent-skills

full-feature-flow

Trigger only when the user explicitly invokes this skill.

Quelltext ansehen

Verwendung

  1. Installationsbefehl kopieren und das Quellpaket vor der Installation in der eigenen Agentenumgebung prüfen.
  2. Benötigte Werkzeuge, Zugangsdaten und Hosts im Original prüfen.
  3. Eingabe und gewünschtes Ergebnis beschreiben und die Ausgabe mit den Quellen abgleichen.
npx skills add https://github.com/sztadii/agent-skills --skill full-feature-flow

Vor dem Start

  • Die Installation umfasst weder Modellabo noch API-Kontingent oder Dateizugriffsrechte.
  • Sterne und Installationen zeigen Aktivität, keine unabhängige Qualitäts- oder Sicherheitsprüfung.
Originales Skill-Dokument

Aus dem Quell-Repository gerendert; Überschriften, Beispiele, Code, Tabellen, Links und Bilder bleiben erhalten.

Full Feature Flow

Scope

Take a feature, fix or skill change from idea to working-tree code: a Planner, helped by two parallel analysts, agrees on the plan with the user, and a Coder implements it and requests verification from a Tester. The Tester manages its agents according to its own instructions and returns a verdict and validated findings. The Coder fixes findings in up to three rounds, runs the final checks, prepares screenshots or recordings of the final version and reports the results. The main agent (you) only orchestrates: it gathers context, relays the Planner's questions and plans, collects the user's approval and starts the Coder. It does not plan, implement or review.

Each role's model, effort, responsibilities and output live in its reference file; use them as written unless the user overrides a setting:

Invocation authorizes planning, editing the working tree and starting the agents above. It does not authorize commits, pushes, PRs or publication; honor narrower user constraints and pass them to every agent.

Workflow

Agents start with no parent history, so every brief must be self-contained: include the task, constraints, relevant context and the absolute path of the role's reference file, and tell the agent to read it first. The Planner also needs the paths of its analysts' files, and the Coder the paths of the Tester's and its agents' files, to pass them down. In T3 Code, check orchestrator_capabilities once per thread and start agents with delegate_task using the provider, model and options from the reference file; follow each task's next without polling.

Plan

  1. Gather context: the user's request, relevant repository instructions, files

and conventions, and known constraints. Resolve verifiable repository facts yourself.

  1. Start the Planner (role: "design", shared checkout) with that context.

It runs its analysts according to its own reference file and returns exactly three questions at once. Do not start the analysts yourself.

  1. Show the user all three questions together and wait for the answers. Send

the answers to the Planner's thread (t3_thread_send to its child thread) so it keeps its context; it returns a plan of at most six concise sentences.

  1. Show the user the complete plan and ask whether they approve it. If not,

send their feedback to the same Planner thread and show the revised complete plan, also limited to six concise sentences. Repeat until the user explicitly approves. Never approve on the user's behalf or treat silence or unrelated replies as approval.

  1. After approval, stop the Planner (task_cancel its task) and keep the

approved plan verbatim. If the user cancels, stop the Planner and end the workflow.

Build and test

  1. Start the Coder (role: "implementation", shared checkout) with the

approved plan verbatim, the gathered context, user constraints and the absolute paths of every reference file except the Planner's and its analysts'.

  1. The Coder implements the plan and prepares access for testing: a running

web app's URL, a built mobile app or an executable command or script for other changes. It then starts a Tester for each round. The Tester manages verification according to its own reference file and returns a verdict and validated findings. The Coder fixes them until a round passes, for at most three rounds. After a passing round or round three, even if findings remain disputed or open, it runs the whole unit suite, e2e tests and build. If checks fail, it fixes them and verifies only the failures first, then reruns the complete final set once those pass. It then captures, or reuses from earlier rounds, a screenshot or 1080p 60 fps recording of the final version and reports the results, media and unresolved disagreements. Do not start the Tester or its agents yourself.

  1. If the Coder returns a blocking question, relay it to the user and send the

answer to the Coder's thread; the Coder starts over with implementation and the Tester rounds. Otherwise wait for its final report.

Result

Relay the Coder's report to the user, leading with whether the requested change was fully implemented and whether any problems remain. Include the prepared app's URL, artifact path or invocation, its readiness, required setup and services left running for the user or a later agent, and the visual evidence's absolute paths, captions, quality and source snapshot or its capture gap, so the user can review it and full-create-pr can attach it. Then give the approved plan summary, changed paths, delivered behavior, checks and results, rounds run with each Tester's agents, user flow results and findings, findings fixed, findings declined with reasons and evidence, unresolved disagreements after the third round, open findings, full test run results and remaining risks. If any agent was unavailable or the workflow stopped, state which stage completed, what blocks the rest and the decision needed. Do not report the task as done unless the Coder reports it done.

aus demselben Repository

Weitere Skills

Alle Skills