AGENT SKILL

atl-develop

Implement or review changes in the ATL repository with the correct architecture owner, generated-tree boundary, risk-classified verification, and privacy checks. Use for ATL code, documentation, tests, build tooling, refactors, or bug fixes. Do not use merely to operate the installed ATL CLI against Jira or Confluence.

License 许可
Apache-2.0
所属仓库
isukharev/atl
最近更新
2026-09-14
Stars
★ 7
Forks
1

01 前言 Front Matter

name
atl-develop
description
Implement or review changes in the ATL repository with the correct architecture owner, generated-tree boundary, risk-classified verification, and privacy checks. Use for ATL code, documentation, tests, build tooling, refactors, or bug fixes. Do not use merely to operate the installed ATL CLI against Jira or Confluence.

02 资源文件 1

03 SKILL.md 正文

Develop ATL

  1. Treat the active repository AGENTS.md instruction chain as binding. Do not reread the whole file when Codex already supplied it in session context unless evidence says the active copy is missing or stale.
  2. Read Development and verification. Apply Efficient agent work before any long command, large read, or repeated status check.
  3. Run the runbook's literal read-only snapshot once per session boundary and classify every dirty path before executing Makefiles/scripts or editing. Do not repeat identity/toolchain probes per commit.
  4. For non-trivial work, require an issue and public agent plan that states the low, standard, or high process class before code.
  5. Find the canonical subsystem and documentation owner. Never edit generated client skill trees directly. Apply Simplify within the contract before implementation and during the existing integrated review.
  6. Iterate with focused local tests and the class-required privacy/review contour. Once stable, use the landing workflow to request impact-selected hosted gates for the exact PR head/base; do not duplicate full suites locally. Raw Go uses the runbook's automatic toolchain; never derive or pin GOROOT.
  7. Use $atl-land-change when a coherent planned slice is ready for a draft PR or when an existing PR needs review, CI, and merge handling.

Stop and report rather than guessing when a dirty path has unknown ownership, the requested scope lacks authority, or the change would cross an architecture, durable-format, security, or live-write boundary not covered by the plan.

04 同仓库更多 Skills

isukharev
isukharev/atl

triage-issue

Triage a bug against Jira with atl before proposing a ticket. USE WHEN an error or stack trace comes with duplicate, prior-fix, or file-a-bug intent. DO NOT USE WHEN debugging code only, inspecting a known issue, or creating an ordinary task.

★ 7
isukharev
isukharev/atl

status-report

Draft an audience-oriented Jira status report with atl. USE WHEN the user wants a weekly/daily progress narrative, blocker summary, or period report. DO NOT USE WHEN the output is a visual dashboard, direct Jira query, knowledge search, or codebase summary.

★ 7
isukharev
isukharev/atl

sprint-dashboard

Build a visual read-only Jira sprint or board dashboard with atl. USE WHEN the user wants a standup view, WIP picture, or sprint snapshot. DO NOT USE WHEN the output is a narrative report, raw Jira list, or codebase dashboard.

★ 7
isukharev
isukharev/atl

spec-to-backlog

Convert a specification into a reviewed Jira epic and backlog with atl. USE WHEN decomposing requirements, RFCs, or designs into linked Jira work. DO NOT USE WHEN the source is meeting notes, only a summary is wanted, or the task is a direct service operation.

★ 7
isukharev
isukharev/atl

setup

Install and configure atl authentication, backends, and mirror defaults. USE WHEN the user explicitly requests install, authentication, setup repair, or /atl:setup. DO NOT USE WHEN handling normal Jira, Confluence, search, reporting, or mirror work; explicit-only.

★ 7
isukharev
isukharev/atl

search-knowledge

Answer from qualified Jira and Confluence evidence with atl. USE WHEN organizational knowledge is unknown or split across both services. DO NOT USE WHEN one exact resource is known, the task is a report/dashboard, or the answer belongs in the codebase.

★ 7