Skip to content

docs: add engineering best-practices (AI readiness) audit - #37

Merged
amitiwari1411 merged 1 commit into
mainfrom
docs/ai-readiness-audit
Sep 11, 2026
Merged

amitiwari1411 merged 1 commit into
mainfrom
docs/ai-readiness-audit

Conversation

@oly-agents

@oly-agents oly-agents Bot commented Sep 11, 2026

Copy link
Copy Markdown
Contributor

Summary

Adds docs/engineering-best-practices-audit.md — an evidence-based AI-readiness audit of this
repo against the 49-item engineering best-practices checklist (rubric item-credit-v1,
2026-09-04) — and indexes it from the README under a new Repo Health section.

This is documentation only. No source, test, CI, or config behavior changes.

Headline results

Metric Value
Critical gates RED
Adjusted compliance 46.3%(14 Met + 0.5 x 9 Partial) / (49 - 9 N/A)
Status counts 14 Met · 9 Partial · 17 Gap · 9 N/A

Critical gates are RED because of three items: AGENTS.md (absent), README setup & run
instructions
(the README is a link index with no description, install, or dev-loop steps), and
required CI checks before merge — six jobs run on every PR, but org ruleset require-pr-review
(id 3174764) carries no required_status_checks rule, so a red build does not block a merge. The
other applicable gates (branch protection, secret scanning, SAST, unit tests, integration tests,
scoped secrets, lockfiles) are Met.

What the audit covers

  • Profile — published Python library (src/ layout, hatchling wheel + git tag release),
    Python 3.10 / uv / ruff / pytest / poe, external surfaces Snowflake, S3, and
    Outerbounds; no deployed runtime, no owned schema, no UI. Ownership verified as patterninc,
    so inherited Wiz and Toolsmith controls apply (Wiz check runs observed on PR docs: document TABLE_OWNERSHIP_REGISTRY as a daily-refreshed table #34's head).
  • Per-item verdicts for all 49 items with concrete evidence (file paths, CI job names,
    ruleset ids, observed check runs).
  • 26 prioritized, ticket-ready recommendations sized S/M/L, critical gates first.
  • 9 declined practices with profile-backed rationales (no service boundary, no UI, no local
    DB, nothing deployed, no owned schema).
  • Beyond the checklist — strengths the rubric does not name, e.g. CI tests the built wheel
    rather than the source tree, and the Check Version job makes a forgotten version bump fail on
    a duplicate tag.

Notes for review

The audit is a snapshot of commit 60ea68e on main; open PRs (#33 mkdocs, #35 agent env config,
#36 Release Please) are referenced as in-flight but not credited as Met. CI does not run on this
PR — the workflow's paths-ignore excludes README.md and docs/**.


Oly run: https://oly.pattern.com/runs/01a08ef6-1ea4-74cc-9aa2-36cc47a9c05d
Requested by: amitkumar.tiwari@pattern.com

Audits the repo against the 49-item engineering best-practices checklist
and records evidence, verdicts, scoring, and prioritized remediation.

Index the report from the README alongside the existing docs.
@amitiwari1411
amitiwari1411 merged commit 51da772 into main Sep 11, 2026
6 checks passed
@amitiwari1411
amitiwari1411 deleted the docs/ai-readiness-audit branch September 11, 2026 06:08
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant