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 pedroknigge/arkgate --skill ark-autopilotgit clone --depth 1 https://github.com/pedroknigge/arkgateWrote 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/pedroknigge/arkgate/ark-autopilot)<a href="https://agentmods.dev/skills/pedroknigge/arkgate/ark-autopilot"><img src="https://agentmods.dev/badge/skills/pedroknigge/arkgate/ark-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/pedroknigge/arkgate/ark-autopilot"><img src="https://agentmods.dev/badge/skills/pedroknigge/arkgate/ark-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 | $0.00024 | $0.05640 |
| Opus 5 | $0.00012 | $0.02820 |
| Sonnet 5 | $0.00005 | $0.01128 |
| Haiku 4.5 | $0.00002 | $0.00564 |
Grade A, and why
ark-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 yesterday.
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 — 302 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ark-autopilot — Apply the path
Contener · Guiar · Ordenar. In plain words: contain the write, guide the next step, order leftover mess. This door is Guiar: apply leftover design and illegal imports, one small change at a time. Skills never enforce — CLI / hooks / CI do.
When: the user wants architecture cleaned end-to-end, or leftover design after a map.
Not when: map only (/ark-explore), session 0 / config lying (/ark-adopt), or one new file (/ark-place).
Steps
- Read
arkgate-check --doctor(do #1). - Fix import-rule debt; if leftover design remains, apply one small refactor.
- Re-run doctor. Never weaken
ark.config.json.
This is the apply door. Invoking it is the approval. Run arkgate-check --doctor for the next step.
Autonomy contract
The CLI is a sensor and gate. CLI budget: one doctor / --plan; open files; write;
re-check. Forbidden as the result: “4 safe / 4 judgment — approve?”, dumping --plan JSON,
or STOP — invoke /ark-fix / /ark-loop / /ark-contract. Do that work here.
Apply plan A (mechanical-safe and judgment). If leftover design work remains, apply
one small refactor this turn, then re-doctor. Never batch every Shape bet.
Composes explore → sensors → dual plan → write. Validate with ark-check; never weaken the gate.
Team lock: do not edit ark.config.json, arkrules/*, or .ark-baseline.json in this
turn unless the user invoked a steward contract session. Feature diffs use
ark-check --changed --base <merge-ref>. Growing the baseline is steward-only.
Not a plan grinder. Empty --plan does not mean “architecture is healthy” without
the explore pass and dual-plan section B (pattern / Shape bets).
Improvement compass (process preflight)
When doctor is available, read doctor.improvementCompass (or the human Improvement compass section).
Name 1–3 residual lenses in plain language before skill-shopping. Always notAScore — never invent
0–10 scores or Excellent/Good ranks.
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.
- yesterday Changed · +4 lines 3473d4e7b25f
- 4d ago Changed 507aafd0fb82
- 8d ago First seen · 298 lines · 24 tokens per session scan A 17502ffa7988
ark-autopilot is a skill published in the GitHub repository pedroknigge/arkgate (1 stars, last pushed yesterday), licensed MIT. It adds 24 tokens to every session and 5,640 once invoked, about $0.0001 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
architecture-patterns
Software architecture patterns and best practices.
architecture-patterns
Software architecture patterns and best practices.
truecourse-fix
Fix TrueCourse violations that have suggested fixes.
truecourse-list
List TrueCourse violations found in this repository.
architecture-paradigm-event-driven
Applies event-driven async messaging to decouple producers and consumers. Use when designing real-time or multi-subscriber systems needing loose coupling.
architecture-paradigm-serverless
Applies serverless FaaS patterns for event-driven workloads. Use when designing bursty workloads with minimal infrastructure and pay-per-execution cost model.