docs(stories): add story 09 — standing audit loops in practice

[fyodoriv]

Why needed

Story 08 lists standing audit loops alongside the rest of the rich-metadata fields, but only sketches the pattern — readers had no full walkthrough to model their own usage on. Story 09 closes that gap with one complete cycle: when to reach for the pattern, the canonical TASKS.md shape, how pickBestTask and /next-task handle it, the audit pass itself, and the anti-pattern to avoid. Closing user-stories-09-standing-loops-walkthrough.

What changed

• New docs/user-stories/09-standing-audit-loops.md with the five sections required by the task brief (When to reach for one, The shape, How agents pick it up, The agent's audit pass, Anti-pattern: don't over-loop) plus the ## Try it yourself demo that PR docs: add Try-it-yourself runnable demo to every user story #60 established as the template for every story.
• docs/user-stories/README.md — new row 9 in the user-stories index table, between row 8 and the Automation Status section.
• docs/user-stories/08-rich-task-metadata.md — closing line of the Standing Audit Loops section now points readers to story 09 for the worked example.
• spec.md — adds the one-line "See User Story 09 for a worked example" cross-link the task brief asked for, with no other spec changes.

Runnable demo

The Try-it-yourself block creates a tmp directory, drops one normal task plus one standing-loop-tagged task into TASKS.md, and shows the contract:

Picked "Document the new pagination flag" (P2)
  Candidates: 1

Candidates: 1 is the proof — the queue has two checkbox tasks, but pickBestTask filtered out the standing-loop one. tasks list --priority P2 then shows both rows (since list enumerates without skipping). I ran the demo end-to-end before committing; the output matches what the doc claims.

Acceptance check

• New story 09 exists with the 5 sections + a Try-it-yourself demo.
• docs/user-stories/README.md table has the new row in the right priority (07 → 08 → 09).
• spec.md standing-loops section gets the one-line "see story 09 for a worked example" link.
• npm run build — passes.
• npm test — 415/415 (cli 183, lint 43, mcp 111, parser 78).
• npm run lint — Checked 1 file(s), found 0 error(s).
• npx -y @tasks-md/lint TASKS.md — Checked 1 file(s), found 0 error(s).

closes user-stories-09-standing-loops-walkthrough

---

🤖 Written by an agent, not Fyodor. Ping me if this looks off.