tracker

Dispatches canonical issue-tracker operations to configured Jira or GitHub Projects adapters.

1|1|Updated Jul 23, 2026
One-click install
npx skills add https://github.com/nearmap/shipyard --skill tracker-nearmap
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tracker
Source: https://github.com/nearmap/shipyard/tree/main/skills/tracker
Command: npx skills add https://github.com/nearmap/shipyard --skill tracker-nearmap

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Workflow skills need to create issues, move them across lifecycle columns, post comments, and attach artifacts without knowing which issue tracker a repository uses. This Skill provides one canonical contract of verbs, statuses, and types, and dispatches every operation to the configured adapter (Jira or GitHub Projects) so core workflows never branch on the tracker. ## Core Features & Use Cases - Canonical tracker contract: A single vocabulary of verbs (create-issue, set-status, post-comment, add-dependency, attach-artifact, and more), five lifecycle statuses, and three issue types that every workflow skill speaks. - Adapter dispatch: Resolves the configured tracker via config, validates required columns and credentials with fail-fast checks, and loads only the matching adapter documentation (Jira REST or GitHub Projects v2 via gh). - Durable conventions: Enforces immutable comments with highest-version execution plans, standalone machine logs via post-log, closure-vs-delivery semantics, and a canonical decomposition procedure. - Use Case: When /sy:ship finishes a task, this Skill moves the issue to done, posts the retrospective comment, attaches the sanitized session transcript, and records ship metrics — against Jira or GitHub without the caller naming either. ## Quick Start Ask the assistant to create a task under the current epic and move it to ready using the configured issue tracker.

Frequently Asked Questions about tracker

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I switch between Jira and GitHub Projects as the issue tracker?▼

Set the tracker key in the repo's .shipyard/config.json to jira or github. The Skill resolves it through the config resolver, validates required columns and adapter configuration, and loads only the matching ADAPTER.md for all tracker mechanics.

How does the Skill map canonical statuses to Jira or GitHub board columns?▼

Five canonical statuses (backlog, ready, in-progress, in-review, done) map to per-repo column names from the columns.* config keys. Jira transitions to the matching target status; GitHub writes the Projects v2 Status single-select field, verified by re-reading the board.

Does the GitHub adapter work with a personal account without an organization?▼

Yes. Type and Status are driven as Projects v2 single-select fields rather than org-only native issue types, so sub-issues, dependencies, comments, and the board all work on GitHub Free for a personal private repo.

Why does preflight fail even when my credentials are set?▼

Presence checks do not prove liveness. The preflight verb performs real authenticated reads — /myself and the project on Jira, or a board read on GitHub — and fails with an Action needed block naming the missing or dead credential without exposing any secret value.

What are the limitations of attaching files to GitHub issues?▼

GitHub issues have no CLI-scriptable attachment, so artifacts upload as private gists linked from a comment. Only UTF-8 text payloads are accepted; non-text payloads are refused outright, whereas Jira's byte-capable attachments accept them once declared opaque.

How are execution plans versioned on an issue?▼

Comments are immutable, so a plan is superseded by posting a new comment with a higher Execution Plan vN heading naming the version it replaces. The highest version number is the plan; position or recency is never used as the selector.