document-it

Document It

Make the next human (or agent) able to use, operate, or change the thing without reverse-engineering — without overbuilding docs.

Operating contract

Shared by every DecisionNerd/dev-skills skill. Canonical copy: handbook/concepts/14-operating-contract.md.

• Define done first. Before using tools, write the finish line in one or two lines: acceptance criteria (existing issue AC, BDD scenarios, tests, or contract; otherwise propose them and where they should live) and the evidence that will prove them. Re-check it before reporting done. Never report done on unverified work.
• Requested scope sets the finish line. A question ("is it ready?", "why is it broken?", "what next?") ends with the answer and a Next: line naming the exact next invoke. An outcome request ("fix", "finish", "land", "#42") continues through the chain (diagnose → fix → test → check-readiness → merge-it) until the outcome or a real blocker. Do not end a turn with "Do you want me to…?" for in-scope, in-repo work.
• Stop only for real blockers. Stop and ask only when you cannot continue without the user, or before: deleting data or unmerged work, force-push or history rewrite, changing anything outside this repository (GitHub objects, deployments, live data, production or paid resources, external services), or leaving the requested scope, unless the user's request already named that exact action. Keep the harness's permission prompts for risky commands. Otherwise keep going and put status notes in the same message as the next action.
• Ask well, once. For a genuine question use the harness's structured question tool when it has one (Claude Code: AskUserQuestion; Codex: request_user_input when the current mode supports it; T3 Code: the question tool) with concrete options; otherwise plain text with numbered options. Treat the answer as settled; do not re-open earlier verdicts, plans, or answers unless asked.
• Fan out when work is parallel. Use subagents for independent reads (repo survey, evidence gathering, per-option research, per-area audits) and independent verification (a reviewer that did not write the change). Writes stay single-owner per path set and sequential. Brief every child with goal, done-when, constraints, must-not, and return shape; verify each result before consolidating. Use Claude Code's Workflow tool only when orchestrating many subagents truly needs it; it is expensive. T3 Code: its delegated subagents — with optional worktree isolation for independent verification — cover this without the Workflow tool.
• Pick the model tier per child; defer to routing config. If the harness or user config already routes subagents, follow it and do not pass a model. Otherwise pick a tier per child: mechanical search or inventory → fast/cheap; implementation and evidence gathering → mid; planning, review, adversarial verification → top. Tier names differ per harness (Claude Code: haiku / sonnet / opus or fable; T3 Code: inherit by default or pick from its live provider/model catalog). Set the choice with the harness's spawn model param or agent frontmatter model: (per-harness details: skills/agents/references/sub.md). Where the harness cannot choose, children inherit or the harness picks; say which in the status note.
• Keep a checklist on long runs. For more than about five steps or work that crosses skills, keep TASKS.md at the repo root and tick items as they finish. Do not commit it unless the repo already tracks one.
• Close every run with three headings. Blocked on me (the one genuine question or blocker, else "none"); Changed (files, commits, GitHub objects, deploys, else "nothing"); Found (evidence, verdict, and Next: <exact invoke>).

Goldilocks first

Don't overuse DocSlime. Init, full-tree fill, or new template pages are earned by missing product-altitude truth — not by "we should document this PR."

When to use

• "Document it" after fix-it / refactor-it / observe-it / ship
• Stale README/runbook/API for the surface you touched
• Real DocSlime gaps (PRODUCT/REQUIREMENTS/TESTING/ADR/…) when the user wants that altitude
• When issues document / pulls document needs a real edit, not a link stub

Workflow

Done when: a skeptic can use the change; commands and links verified against current code.

1. Audience & job

• Who must succeed (user, integrator, operator, developer, agent)?
• Smallest doc change that enables that job?

1. Pick the smallest home

• Prefer an existing page next to the work (README section, runbook, module doc, changelog).
• If a DocSlime tree already exists and the change touches a contract there (requirement, scenario map, obs signal, architecture), update that file surgically — don't open a fill interview for one sentence.
• Only propose docslime init / docslime add when durable product docs are missing and the user wants that system — not as a side effect of documenting today's fix.

1. Write or update

• Match local voice; document commands, contracts, env vars, failure modes, examples that match current code.
• Cut stale claims; don't document aspirations as fact.
• No secrets.

1. Verify

• Commands/paths work (or mark TBD); links resolve.

1. Wrap up

• Next: pulls create / merge-it to land, or point the issue at new paths.
• Next: DocSlime fill/ADR/kiss only if a real product-docs gap remains.

When DocSlime is the right altitude

Use DocSlime structures and methods (not a bigger process than needed):

docs/
├── PRODUCT.md / DESIGN.md / REQUIREMENTS.md
├── experience/ …
└── engineering/ ARCHITECTURE · TESTING · PUBLISHING · OBSERVABILITY · adrs/

Methods when working at that altitude:

• Don't invent product facts; interview only when filling gaps that matter
• Requirements testable/solution-neutral; TESTING maps GWT → evidence
• Hard decisions → ADR (docslime add adr); don't bury them in chat
• Don't keep empty theater templates; don't fork PRODUCT into README
• Finished scaffold pages: remove <!-- LLM: ... --> / italic prompts

Tooling when earned: docslime-install → init / add / fill / adr / kiss.

Guardrails

• Prefer surgical diffs over new trees.
• Don't create parallel taxonomies beside an existing DocSlime docs/.
• Don't init DocSlime "while we're here" after a small code change.
• Keep diffs reviewable.
• Lying docs are still bugs — fix the claim you invalidated; don't boil the ocean.

Output

**Audience / job**
- ...

**Scope**
- surgical | docslime-altitude
- why this altitude: ...

**Docs changed**
- paths: ...

**What a reader can now do**
- ...

## Blocked on me

none

## Changed

- ...

## Found

- ...
- Next: pulls create / merge-it

Grounding

Diátaxis / Write the Docs: accuracy over volume. DocSlime is the house product-docs system when altitude warrants it (handbook/concepts/13-quality-trace.md). KISS: don't overbuild (handbook/practices/kiss.md). Lying docs are bugs (12-bugs-and-debt.md).

Handbook card: handbook/practices/document-it.md.