Tracer Setup

Setup Tracer

Scaffold the per-repo configuration the tracer skills read:

• Issue tracker — where /tracer-wayfinder charts decision maps, /tracer-to-spec and /tracer-to-tickets publish, and /tracer-code-review fetches specs from (GitHub, GitLab, local markdown, or a workflow you describe)
• Domain docs — where CONTEXT.md and ADRs live, and the consumer rules for reading them

Skipping setup is fine: every skill falls back to local markdown under .tracer/. Run this when you want a real tracker, or to make the conventions explicit for other agents working in the repo.

This is a prompt-driven skill, not a deterministic script. Explore, present what you found, confirm with the user, then write.

Process

1. Explore

Read whatever exists; don't assume:

• git remote -v — GitHub? GitLab? No remote?
• CLAUDE.md / AGENTS.md at the repo root — does either exist? Already an ## Agent skills section?
• docs/agents/ — does this skill's prior output already exist?
• CONTEXT.md, CONTEXT-MAP.md, docs/adr/ — existing domain docs
• .tracer/ — sign the local-markdown convention is already in use
• Monorepo signals — pnpm-workspace.yaml, a workspaces field, a populated packages/* with its own src/. Absence means single-context, which is almost every repo.

2. Present findings and ask

Lead each section with the recommended answer so the user can accept it in a word.

Section A — Issue tracker. If the remote points at GitHub, propose GitHub; GitLab host, propose GitLab; otherwise propose local markdown. The options:

• GitHub — issues in GitHub Issues (gh CLI) → seed from issue-tracker-github.md
• GitLab — issues in GitLab Issues (glab CLI) → seed from issue-tracker-gitlab.md
• Local markdown — files under .tracer/ (projects with no remote) → seed from issue-tracker-local.md
• Other (Jira, Linear, …) — ask the user to describe the workflow in one paragraph; write docs/agents/issue-tracker.md from scratch around it. It must answer how to create and fetch an implementation ticket, express a blocking edge, and mark a ticket agent-ready. It must also include Wayfinding operations: create/update a map, create and parent a decision ticket, query the map-scoped frontier, let one coordinator claim tickets before parallel dispatch, let workers resolve only their child ticket, and let the coordinator apply map-index updates serially.

Section B — Domain docs. Default to single-context — one CONTEXT.md + docs/adr/ at the repo root; write it without asking. Offer multi-context (a root CONTEXT-MAP.md pointing at per-context CONTEXT.md files) only when exploration found monorepo signals. Seed from domain.md.

3. Confirm and write

Show the user drafts of docs/agents/issue-tracker.md, docs/agents/domain.md, and the ## Agent skills block below; let them edit before writing. When updating an existing tracker doc, preserve user customizations and add or refresh its Wayfinding operations rather than replacing the whole file. If the user chose local markdown, verify .tracer/ is ignored and add /.tracer/ to the root .gitignore when needed.

Pick the file for the block: edit CLAUDE.md if it exists, else AGENTS.md; if neither exists, ask which to create. Never create one when the other already exists. If an ## Agent skills block already exists, update it in place.

## Agent skills

### Issue tracker

[one-line summary of where issues are tracked]. See `docs/agents/issue-tracker.md`.

### Domain docs

[one-line summary — "single-context" or "multi-context"]. See `docs/agents/domain.md`.

4. Done

Tell the user which skills now read these files (/tracer-wayfinder, /tracer-to-spec, /tracer-to-tickets, /tracer-code-review, /tracer-implement; all skills for domain docs) and that editing docs/agents/*.md directly is the normal way to adjust later. Re-run this skill when switching trackers or when an older tracker doc needs the Wayfinding operations added.