ad-deepen

A guided architecture review that looks for modules with unclear boundaries or too much responsibility, using terms such as interface, adapter, seam, and locality.

In plain words
What is it for?
Use it in mature team codebases to consider deeper module design improvements. It is not intended for one-line fixes or small experiments.
Why use it?
It helps reveal code structure that makes changes harder than necessary and turns possible improvements into concrete candidates for discussion.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/corridortech/posecap/ad-deepen
Any agent
npx skills add CorridorTech/PoseCap --skill ad-deepen
Clone the repo
git clone --depth 1 https://github.com/CorridorTech/PoseCap

Made for: Claude Code, Codex.

Per session 164 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,873 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00164 $0.01873
Opus 5 $0.00082 $0.00937
Sonnet 5 $0.00033 $0.00375
Haiku 4.5 $0.00016 $0.00187

Measured 2d ago against content hash 1d9839249868, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ad-deepen scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/ad-deepen/SKILL.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

<background_information> Implements ADR-0020 (doc/adr/0020-deep-modules-vocabulary.md) — operational counterpart to the WORKFLOW §8 architectural vocabulary. Surfaces deepening opportunities and walks the user through them. Process scaffold; no primary file output.

Profile scope: team and mature only. Architectural deepening is premature for ≤200-line experiments. poc and solo users do not have this skill installed; to enable deepening, re-init the project at team profile (agentic profile set team) once the codebase has stabilized enough to warrant it. Per ADR-0020 §4.

Codex auto-trigger on description keywords is less mature than Claude Code's. If auto-invocation does not fire when the user mentions deepening, shallow modules, the deletion test, or the two-adapters rule, invoke this skill manually. </background_information>

Route elsewhere when:

  • The change is a one-line fix or mechanical refactor → no scaffold needed.
  • The codebase is poc-shaped (one-shot script, ≤200 lines, no callers): deepening is premature; ship the experiment first.
  • The friction is a bug, not a shapead-diagnose (WORKFLOW §15).
  • The friction is a naming/vocabulary drift — ad-domain.
  • The friction is missing context, not module shape — ad-grill-me.

Step 1 — explore organically. Before naming candidates, read.

  1. Read CONTEXT.md if it exists. Anchor domain vocabulary; deepening proposals must use Customer, Order, Triage role — not Service, Handler, Manager.
  2. Read ARCHITECTURE.md if it exists. Read accepted ADRs in doc/adr/ covering the surface you are about to walk.
  3. Walk the codebase noting friction:
    • Concepts bouncing between modules (locality fault).
    • Interfaces as wide as their implementations (shallow modules — apply the deletion test).
    • Tests that mock four things to exercise one path (test-surface fault).
    • Adapters with one concrete implementation (two-adapters-rule violation, if no second is planned).

Read with eyes for shape, not bugs. The friction is in module geometry; the bug-finding skill is ad-diagnose.

Step 2 — present candidates numbered. For each deepening opportunity, produce a short numbered entry:

### Candidate N: <one-line title using domain + architectural vocabulary>

**Files involved:** `src/foo.ts:42`, `src/bar.ts`.

**Friction:** <one-paragraph plain English — what hurts today, who feels it, when>.

**Proposal:** <one-paragraph plain English — what changes about the module shape>.

**Vocabulary check:** <Module / Interface / Seam / Adapter / Depth / Leverage / Locality applied here, per WORKFLOW §8>.

**Deletion test:** <if you imagine deleting the module today, what happens? Complexity vanishes → delete instead of refactoring. Complexity reappears across N callers → it earns its keep, deepen it>.

**Two-adapters check:** <if introducing a seam: name the second concrete adapter that justifies it. If only one adapter exists or is planned, the seam is hypothetical — drop it from the proposal>.

**Test surface impact:** <does the new shape make the right pattern testable at the call site? If not, the proposal is wrong-shaped>.

**ADR conflicts:** <name any accepted ADR this proposal contradicts. Flag for reopening only if the friction warrants it; otherwise the candidate is rejected>.

Number candidates 1, 2, 3, ... Order by leverage × locality (largest impact first, lowest blast radius among ties). Cap at five.

Read the full file on GitHub · 113 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 113 lines · 164 tokens per session scan A 1d9839249868

Subscribe to this mod's changes

ad-deepen is a skill published in the GitHub repository CorridorTech/PoseCap (190 stars, last pushed 10d ago), licensed Apache-2.0. It adds 164 tokens to every session and 1,873 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens