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.
npx skills add zstevovich/claude-apd --skill apd-pipeline-guidegit clone --depth 1 https://github.com/zstevovich/claude-apdWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zstevovich/claude-apd/apd-pipeline-guide)<a href="https://agentmods.dev/skills/zstevovich/claude-apd/apd-pipeline-guide"><img src="https://agentmods.dev/badge/skills/zstevovich/claude-apd/apd-pipeline-guide/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zstevovich/claude-apd/apd-pipeline-guide"><img src="https://agentmods.dev/badge/skills/zstevovich/claude-apd/apd-pipeline-guide.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00130 | $0.03782 |
| Opus 5 | $0.00065 | $0.01891 |
| Sonnet 5 | $0.00026 | $0.00756 |
| Haiku 4.5 | $0.00013 | $0.00378 |
Grade A, and why
apd-pipeline-guide 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.
APD Pipeline Guide (Codex)
The operating manual for one pipeline cycle. Read it, acknowledge the contract, write the marker, proceed. ~2 minutes; cheaper than any single BLOCK it prevents.
This skill is NOT a clarification dialogue. If the task scope is vague, load
apd-brainstorm FIRST (interactive, optional), converge on a design, then come
back here. This guide is unconditional either way.
Pipeline phase map
spec → builder → reviewer → adversarial → verifier → commit
| Advance | Gate checks at that point |
|---|---|
apd:apd_advance_pipeline('spec', '<task>') |
spec-card.md exists, ≤7 R-criteria, .guide-marker present (this skill) |
apd:apd_advance_pipeline('builder') |
implementation-plan.md exists, plan-spec consistency (strict), regression surface (Cover/Evidence), no stale pre-spec dispatch |
apd:apd_advance_pipeline('reviewer') |
builder ran post-spec, builder cycle cap (default 2) |
apd:apd_adversarial_pass(...) |
only AFTER reviewer.done — out-of-order verdict is refused; and since v6.36 a real native subagent must be on record (apd_prepare_dispatch + spawn_agent), not an inline verdict |
apd:apd_advance_pipeline('verifier') |
.adversarial-summary + .adversarial-rationale.md present, rationale gate, spec-hash immutability |
| commit | guard-git: pipeline complete, commit message prefix, no mass staging |
Two independent spec-card switches, routinely confused:
pipeline_mode: polish— lowers the builder AND reviewer caps 2 → 1. It does NOT skip adversarial: the full builder → reviewer → adversarial → verifier sequence still runs, just once through.adversarial: skip — <reason>— the Lean opt-out, and the ONLY way to skip adversarial. Honoured only at ≤2 R-criteria; at 3+ the opt-out is DENIED (warning) and adversarial stays required at the verifier.
Lean vs Full is declared in the spec; this guide applies to BOTH.
Note: the v6.30 supervision layer (frontier review of the FINAL diff) is
CC-owned and honest-inert on the Codex runtime since v6.33 — a Codex run
structurally cannot dispatch the CC supervisor, so the gate never passes and
never blocks; it logs supervision-not-applicable|runtime=codex and moves on.
This holds on hybrid CC+Codex projects too, even though they share the APD
config and a MODEL_PROFILE declared from the CC side. Do NOT hand-write
.supervision-summary to satisfy it: before v6.33 that produced a FALSE
supervision-pass with no supervisor on record, which is why the gate now
keys on the runtime instead of on the file.
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.
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.
- 11d ago First seen · 256 lines · 130 tokens per session scan A fcee5b2df61a
apd-pipeline-guide is a skill published in the GitHub repository zstevovich/claude-apd (5 stars, last pushed 1mo ago), licensed MIT. It adds 130 tokens to every session and 3,782 once invoked, about $0.0006 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…