agentleFS
ai-agent-assemblyai-agent-assembly/agent-assembly/.claude/skills/jira-delivery/SKILL.md

jira-delivery

Skill4 starsChanged 3 months agoApache-2.0View on GitHub

Carry one Jira ticket through implementation to a merged, Jira-truthful state, following Horonom's company-wide git/PR/merge/Jira invariants — without re-deriving those invariants from scratch each time. Auto-used. Invoke when picking up a Horonom Jira ticket for implementation. Any ticket in the HORO project (or a product project that has adopted this skill) that will produce a code or doc change landing via a PR. - A repo/product with stricter delivery semantics than this skill encodes (e.g. a mandatory second reviewer, a…

The file

Files
SKILL.md
<!-- horonom:generated --><!-- Source: horonomy/.github agents/skills/jira-delivery/SKILL.md. Do not hand-edit — rerun `python3 agents/common/project_skills.py`. -->

SKILL.md — jira-delivery

Purpose

Carry one Jira ticket through implementation to a merged, Jira-truthful state, following Horonom's company-wide git/PR/merge/Jira invariants — without re-deriving those invariants from scratch each time.

Type

Auto-used. Invoke when picking up a Horonom Jira ticket for implementation.

When to use

Any ticket in the HORO project (or a product project that has adopted this skill) that will produce a code or doc change landing via a PR.

When NOT to use

  • A repo/product with stricter delivery semantics than this skill encodes (e.g. a mandatory second reviewer, a release-train merge window, a product-specific pre-merge gate). This skill is a floor, not a ceiling — follow the repo's own CONTRIBUTING.md/.claude/CLAUDE.md wherever it adds a stricter requirement (Company → Product → Repository precedence, see governance/README.md).
  • Work with no Jira ticket and no tracking need (a truly trivial, same-session fix) — don't manufacture ticket overhead where the company's own quick-fix path already applies.
  • A third-party (non-Horonom-owned) repository — never open an issue, PR, or comment there without explicit owner approval; read-only research is fine.

Procedure

  1. Reconcile — read the ticket's full description/AC/comments. Check for prior partial work (comments, existing branches/PRs) before starting.
  2. Worktree — verify the canonical remote and default branch (never assume origin or main); create an isolated worktree per governance/engineering/git-pr-merge.md.
  3. Implement — very small, single-purpose GitEmoji commits (<emoji> (<scope>): <imperative summary>), one logical change each.
  4. Validate — run the repo's actual test/lint/type/build commands locally; a genuinely green local run is authoritative if CI is CI_UNAVAILABLE_EXTERNAL (see governance/workspace/ci-classification.md) — never a substitute for a real, reproducible CI_FAILED.
  5. PR — title [<ticket>] <emoji> (<scope>): <summary>, body follows the repo's PR template. Never merge directly to the default branch.
  6. Review — run the full pre-merge checklist in governance/engineering/git-pr-merge.md (diff vs. AC, correctness/ design/security, tests, CI, screenshot evidence for UI changes). An implementation sub-agent never merges its own PR.
  7. Merge — "Create a merge commit" only; never squash or rebase-merge.
  8. Jira update — comment with concrete evidence (PR link, merge commit, validation commands/results), then transition to Done only when the AC is genuinely satisfied — never on status alone.
  9. Clean up — remove the worktree and, once merged, the branch (local and remote).

References

  • governance/engineering/git-pr-merge.md — the full git/PR/merge rules this procedure implements.
  • governance/engineering/jira-delivery.md — the Jira-specific invariants (project/hierarchy, truthful status, external-repo restriction).
  • governance/engineering/testing-review.md — testing and review invariants.
  • governance/workspace/ci-classification.md — CI_FAILED vs. CI_UNAVAILABLE_EXTERNAL.

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.