product-audit

Name: product-audit
Author: gtrabanco

36installs2forks

How to install product-audit?

gtrabanco/agentic-workflow/product-audit

$npx -y skills add gtrabanco/agentic-workflow --skill product-audit

Installs into the current project.

Summary

Periodic, product-wide health check — the CTO's "where do we actually stand?" Sweeps the WHOLE codebase (not a diff, not a PR) across every applicable axis — correctness, architecture, security/cybersecurity, performance, tests, UX/UI, accessibility, SEO, brand, tech debt — PLUS

Files · 1

View on GitHub

SKILL.md

1	# Product Audit
2
3	The CTO health check: run every few features, before a release, or when the
4	product is "done", to answer *"where do we actually stand, and what should we do
5	next?"* across the entire product. **Recommend-only — it never fixes, opens
6	issues, or edits the roadmap. It proposes; the human decides.** Its single
7	write is its own persisted report: `docs/audits/<n>-<YYYY-MM-DD>.md`.
8
9	## Turn contract — verify before ending the turn
10
11	```
12	✓ The full PRODUCT AUDIT report was printed in the fixed output format (health by dimension, F-numbered ranked findings, four proposal streams — the roadmap streams always present, `none — <why>` when empty)
13	✓ The report was WRITTEN to `docs/audits/<n>-<YYYY-MM-DD>.md` (incremental audit id) and committed — the file is the durable deliverable, the chat print is a copy
14	✓ Nothing else was fixed, filed, or changed — the report file is the ONLY mutation
15	✓ The closing `→ Next:` block is printed as the ABSOLUTE last output
16	```
17
18	About to end the turn with any box unchecked? The turn is NOT done — complete
19	the missing box first (weak models drop end-of-document duties; this list is
20	first on purpose).
21
22	## When to use
23
24	- Periodically (every few features) or at a product-ready milestone.
25	- When you want the broad, honest picture — quality, security, debt, docs, and
26	roadmap — not the review of a single change (`review-change`) or PR (`audit-pr`).
27
28	This is the widest lens in the workflow. `review-change` audits a diff, `audit-pr`
29	a PR, `audit-docs` doc↔roadmap↔code coherence — `product-audit` audits the
30	whole product across every dimension and turns what it finds into proposals.
31
32	## Scope
33
34	The entire codebase and its process artifacts: source, tests, the docs tree, the
35	roadmap, the fix index, open issues, and every feature folder's planning docs.
36	Accept an optional path/area to focus a partial audit; state the scope and, if you
37	sample rather than exhaust a dimension, say what you sampled — never imply full
38	coverage you didn't do.
39
40	> Tip (provisional). For the broadest, deepest run, the user can turn on
41	> `ultracode` (`/effort ultracode` — a Claude Code session setting pairing xhigh
42	> effort with automatic multi-agent orchestration) so this sweep fans out across
43	> parallel subagents instead of one context window. It's a research-preview feature
44	> and a session choice — not something this skill declares (no skill can set
45	> `effort: ultracode`). On agents without it, run the audit as-is: sequential
46	> passes over each dimension — only wall-clock changes, never coverage.
47
48	## Step 0 — Discover the project (always first)
49
50	Per the agent guide's Workflow conventions + documentation map, then read
51	what THIS skill needs: the roadmap, the fix index, the feature folder layout, and
52	the verification gate. From the map decide the product's nature (web / mobile /
53	console / library / backend / infra) and which axes apply — the same applicability
54	logic `review-change` uses, applied product-wide. Note any optional platform
55	review skills the project installed (extras, never requirements — the internal
56	pack covers every axis).
57
58	## Audit dimensions (platform-adaptive — run only what applies)
59
60	\| Dimension \| What it sweeps product-wide \| Applies to \|
61	\|---\|---\|---\|
62	\| Correctness & architecture \| Bugs, layer/boundary violations, dead code, overengineering, drift from the architecture doc \| all \|
63	\| Security & cybersecurity \| Secrets in repo, authz gaps, input validation, depe

People who liked this also viewed

6 picks

Type

product-audit

36installs2forks

View on GitHub

How to install product-audit?

gtrabanco/agentic-workflow/product-audit

$npx -y skills add gtrabanco/agentic-workflow --skill product-audit

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Product Audit
2
3	The CTO health check: run every few features, before a release, or when the
4	product is "done", to answer *"where do we actually stand, and what should we do
5	next?"* across the entire product. **Recommend-only — it never fixes, opens
6	issues, or edits the roadmap. It proposes; the human decides.** Its single
7	write is its own persisted report: `docs/audits/<n>-<YYYY-MM-DD>.md`.
8
9	## Turn contract — verify before ending the turn
10
11	```
12	✓ The full PRODUCT AUDIT report was printed in the fixed output format (health by dimension, F-numbered ranked findings, four proposal streams — the roadmap streams always present, `none — <why>` when empty)
13	✓ The report was WRITTEN to `docs/audits/<n>-<YYYY-MM-DD>.md` (incremental audit id) and committed — the file is the durable deliverable, the chat print is a copy
14	✓ Nothing else was fixed, filed, or changed — the report file is the ONLY mutation
15	✓ The closing `→ Next:` block is printed as the ABSOLUTE last output
16	```
17
18	About to end the turn with any box unchecked? The turn is NOT done — complete
19	the missing box first (weak models drop end-of-document duties; this list is
20	first on purpose).
21
22	## When to use
23
24	- Periodically (every few features) or at a product-ready milestone.
25	- When you want the broad, honest picture — quality, security, debt, docs, and
26	roadmap — not the review of a single change (`review-change`) or PR (`audit-pr`).
27
28	This is the widest lens in the workflow. `review-change` audits a diff, `audit-pr`
29	a PR, `audit-docs` doc↔roadmap↔code coherence — `product-audit` audits the
30	whole product across every dimension and turns what it finds into proposals.
31
32	## Scope
33
34	The entire codebase and its process artifacts: source, tests, the docs tree, the
35	roadmap, the fix index, open issues, and every feature folder's planning docs.
36	Accept an optional path/area to focus a partial audit; state the scope and, if you
37	sample rather than exhaust a dimension, say what you sampled — never imply full
38	coverage you didn't do.
39
40	> Tip (provisional). For the broadest, deepest run, the user can turn on
41	> `ultracode` (`/effort ultracode` — a Claude Code session setting pairing xhigh
42	> effort with automatic multi-agent orchestration) so this sweep fans out across
43	> parallel subagents instead of one context window. It's a research-preview feature
44	> and a session choice — not something this skill declares (no skill can set
45	> `effort: ultracode`). On agents without it, run the audit as-is: sequential
46	> passes over each dimension — only wall-clock changes, never coverage.
47
48	## Step 0 — Discover the project (always first)
49
50	Per the agent guide's Workflow conventions + documentation map, then read
51	what THIS skill needs: the roadmap, the fix index, the feature folder layout, and
52	the verification gate. From the map decide the product's nature (web / mobile /
53	console / library / backend / infra) and which axes apply — the same applicability
54	logic `review-change` uses, applied product-wide. Note any optional platform
55	review skills the project installed (extras, never requirements — the internal
56	pack covers every axis).
57
58	## Audit dimensions (platform-adaptive — run only what applies)
59
60	\| Dimension \| What it sweeps product-wide \| Applies to \|
61	\|---\|---\|---\|
62	\| Correctness & architecture \| Bugs, layer/boundary violations, dead code, overengineering, drift from the architecture doc \| all \|
63	\| Security & cybersecurity \| Secrets in repo, authz gaps, input validation, depe

People who liked this also viewed

6 picks

Type