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 ahmedawan-oracle/claude-code-plugins --skill aidp-fusion-autopilotgit clone --depth 1 https://github.com/ahmedawan-oracle/claude-code-pluginsWrote 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/ahmedawan-oracle/claude-code-plugins/aidp-fusion-autopilot)<a href="https://agentmods.dev/skills/ahmedawan-oracle/claude-code-plugins/aidp-fusion-autopilot"><img src="https://agentmods.dev/badge/skills/ahmedawan-oracle/claude-code-plugins/aidp-fusion-autopilot/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/ahmedawan-oracle/claude-code-plugins/aidp-fusion-autopilot"><img src="https://agentmods.dev/badge/skills/ahmedawan-oracle/claude-code-plugins/aidp-fusion-autopilot.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 | $0.00234 | $0.04198 |
| Opus 5 | $0.00117 | $0.02099 |
| Sonnet 5 | $0.00047 | $0.00840 |
| Haiku 4.5 | $0.00023 | $0.00420 |
Grade A, and why
aidp-fusion-autopilot 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 10d 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 — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.
aidp-fusion-autopilot — one goal, driven A-to-Z
Turns "I want a CFO dashboard of supplier spend vs GL balance, by currency" into a finished workbook by conducting the existing skill family — it does not re-implement any step. Its only jobs are: detect where the user already is, drive the next incomplete step via the right skill, and stop at the decisions a human must make.
This is the single entry point so the user never has to know which of the seven skills to invoke, or in what order.
When to use
- The user states a goal and wants it taken end-to-end ("take me from nothing to a dashboard", "set up Fusion analytics for X", "autopilot this").
- Resuming a half-done journey ("finish setting up my supplier dashboard").
When NOT to use
- A single known step → call that skill directly (
/aidp-fusion-seed,/oac-dataset-advisor,/mart-author,/workbook-authoring, …). Autopilot is overhead when the user already knows the one thing they want.
Operating principles
- Compose, never reimplement. Each step delegates to a sibling skill or a CLI command. Autopilot owns only sequencing + state.
- State-first — don't redo finished work. Detect each step's status before acting (a tenant with gold already seeded skips straight to advise).
- Inherit the sub-skills' guards. The seed destructive guard, the advisor's live-evidence rule, bootstrap's variation-freeze surfacing — autopilot never weakens them. It auto-advances only on a clean result.
- Pause at human decisions (see gates). Auto-advance otherwise. Always show the journey state + what it's about to do before a state-changing step.
The journey (step state machine)
| # | Step | Done when (detect) | Drive with | PAUSE before if |
|---|---|---|---|---|
| 1 | Config | bundle.yaml + aidp.config.yaml exist; coords non-placeholder |
aidp-fusion-autopilot init (scaffold if absent — fresh install) → /aidp-fusion-config for coords |
missing fusion: connectivity (human-only) |
| 1b | OAC MCP connect (front-loaded prerequisite — see §below) | oac-mcp-server tools answer a live search_catalog ping |
project-scoped dashboard mcp-setup, then restart/reconnect Claude Code |
always when dead — staging the connector needs a Claude Code restart before its tools work; write the resume checkpoint, then PAUSE for the restart |
| 2 | Bootstrap | profiles/<tenant>.yaml present + fingerprint pinned |
/aidp-fusion-bootstrap |
multi-match variation needs a human pick (never --non-interactive); surface frozen picks |
| 3 | Seed | live gold has the needed tables (probe) | /aidp-fusion-seed |
always confirm the destructive guard's CONFIRM outcome |
| 4 | Advise | — (always run for the goal) | /oac-dataset-advisor |
— |
| 5 | Author mart (only on advisor GAP) | the gap node exists live after seed | /mart-author → use-pack → back to Step 3 |
confirm the authored change before seeding it |
| 6 | OAC dataset | a dataset over the recommended table(s) exists (OAC MCP) | /oac-dataset-setup |
always — dataset creation is an OAC UI action today (MCP can't create datasets); hand the exact spec to the user |
| 7 | Workbook | a workbook on that dataset exists/renders | /workbook-authoring |
confirm overwrite if replacing an existing workbook |
| 8 | End-user MCP chat (optional deliverable) | least-privilege OAC user handed the connect steps | docs/oac_mcp_setup.md hand-off |
confirm the end-user OAC account is least-privilege (v1.4 exposes write/delete/ACL tools) |
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.
- 10d ago First seen · 259 lines · 234 tokens per session scan A 6177b3ce1e5f
aidp-fusion-autopilot is a skill published in the GitHub repository ahmedawan-oracle/claude-code-plugins (2 stars, last pushed 1mo ago), licensed MIT. It adds 234 tokens to every session and 4,198 once invoked, about $0.0012 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…