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 arcayne/agentic-delivery-playbook --skill pigit clone --depth 1 https://github.com/arcayne/agentic-delivery-playbookWrote 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/arcayne/agentic-delivery-playbook/pi)<a href="https://agentmods.dev/skills/arcayne/agentic-delivery-playbook/pi"><img src="https://agentmods.dev/badge/skills/arcayne/agentic-delivery-playbook/pi/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/arcayne/agentic-delivery-playbook/pi"><img src="https://agentmods.dev/badge/skills/arcayne/agentic-delivery-playbook/pi.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 | $0.00057 | $0.00653 |
| Opus 5.5 | $0.00023 | $0.00261 |
| Sonnet 5.5 | $0.00011 | $0.00131 |
| Haiku 4.5 | $0.00006 | $0.00065 |
Grade A, and why
agentic-delivery-playbook 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 1mo 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 — 28 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agentic Delivery Playbook for Pi
Canonical policy
- Read
../../playbook.mdfor the Direct and Controlled process rules. - Read
../../profiles/gpt-6.mdfor task-fit routes and client limits. - Use
../../templates/contract.mdand../../templates/run.jsonwhen the task warrants durable evidence; do not add ceremony to clear, low-risk work. - Keep Pi as an adapter to the same product. Do not introduce a separate Pi-only process mode such as Lightweight or Full.
Pi-specific route verification
A Pi package install makes this skill available; it does not set model, effort, worker, or reviewer defaults. This repository's optional project profile in .pi/settings.json configures fresh trusted project sessions to start the parent/orchestrator at openai-codex/gpt-6-sol / medium, the implementation worker at openai-codex/gpt-6-luna / high, and the reviewer at openai-codex/gpt-6-sol / medium. The matching template is optional and is not applied by package installation. Project settings affect startup only: trust is required, existing sessions do not change, and CLI/SDK overrides or manual selection may override them.
For this profile, the Sol Medium parent owns planning, contract and task launch, sequencing, validation, synthesis, and continuation. The explicitly routed Luna High worker is the sole writer for every implementation edit, including small edits. A fresh-context Sol Medium reviewer is read-only and checks the actual diff and evidence; reviewer identity alone does not establish independence. Keep three facts distinct: requested route is the task instruction, configured route is settings intent, and observed route is the effective runtime route. Verify the effective Pi version, visible tools, model/effort, and role route independently; settings establish configuration capability, not route attestation.
Use only capabilities exposed in the current session. If a required route is unavailable or ineffective, stop that affected lane and ask for approval; do not silently substitute or claim delegation or independent review. Record unexposed route details as unknown/runtime-default. A missing project settings file alone is not a package failure, but does not meet this profile's requested startup configuration.
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.
- 1mo ago Changed · +2 lines 3150562817b3
- 1mo ago Changed · -522 lines · -20 tokens per session 0ebc7c252d7d
- 1mo ago First seen · 548 lines · 77 tokens per session scan A c3bc83c27ddf
agentic-delivery-playbook is a skill published in the GitHub repository arcayne/agentic-delivery-playbook (10 stars, last pushed 8d ago), licensed MIT. It adds 57 tokens to every session and 653 once invoked, about $0.0002 per session on Opus 5.5. 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
owasp-reviewer-prompt
A paranoid OWASP-Top-10-aware system prompt for AI code review that traces data flow, treats every input as malicious, maps each finding to an OWASP category, and outputs a structured Summary / Findings / Severity / Score block reviewers can act on.
lightweight-monitoring
Add lightweight application monitoring, health checks, and request metrics with minimal dependencies and operational overhead. Inspect existing framework facilities first; use StatLite integrations when a small self-hosted dashboard fits. Use for small applications and VPS deployments, not broad observability…
beautify-with-pingfusi
Beautify or redesign an existing website through iterative pingfusi review rounds with a real human reviewer. Use when asked to "beautify this website," "make this page look professional," "polish this UI/design," "improve the visual design," or finish an AI-built page when there is no reference site to match. Do not…
review-video-with-pingfusi
Have any video reviewed by a real human, through iterative pingfusi review rounds. Use when asked to "review this video", "check the rendered video", "does this video match the prompt/brief", "what do people think of this ad/trailer/demo", or after rendering a Remotion composition or AI-generated clip that no test can…
sdlc-regulatory
AI-driven SDLC governance framework with quality gates, risk-based autonomy caps, change traceability, Definition of Done, and a compliance crosswalk (EU AI Act, NIST AI RMF, ISO/IEC 42001, SOC 2, GDPR). Use when running or reviewing work through the BMad-style subagent SDLC, enforcing gates, classifying risk, or…
article-polish
Review articles for readability, clarity, narrative flow, repetition, unnatural phrasing, and reader friction, including screenshots, charts, captions, and labels. Use for requests to polish an article, review a draft for readability, improve article prose, or check writing before publication across genres. Preserve…