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 agentmods add skills/pedroknigge/arkgate/ark-loopnpx skills add pedroknigge/arkgate --skill ark-loopgit clone --depth 1 https://github.com/pedroknigge/arkgateWhat 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 | $0.00022 | $0.01099 |
| Opus 5 | $0.00011 | $0.00549 |
| Sonnet 5 | $0.00004 | $0.00220 |
| Haiku 4.5 | $0.00002 | $0.00110 |
Grade A, and why
ark-loop 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.
How it starts
The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ark-loop — Shortcut to /ark-autopilot
Not a first-run door. This leftover name is a shortcut. Driving --plan to goal.met is /ark-autopilot.
Do that job now. Auto-apply only the four mechanical-safe kinds; judgment you write.
Autonomy contract
Invoking this leftover name is the approval to apply plan A. Open every step file.
Write. Re-check. Empty plan A + leftover design → /ark-explore then /ark-autopilot
for one extraction card (never mechanical-safe B).
When / not when
| Use this leftover name when… | Prefer instead |
|---|---|
Plan A has steps; drive to goal.met |
/ark-autopilot |
| Mechanical-safe + judgment apply | Map only → /ark-explore; session 0 → /ark-adopt |
Dual engine (mandatory)
| Engine | Role |
|---|---|
| Deterministic | Only the four mechanical-safe kinds auto-apply; plan tags; gate re-check |
| Exploratory | Read sources; detect concentrated edges / false-green before grinding |
The CLI is a sensor, never the whole job. Claiming done without the exploratory bar for this skill is incomplete.
MCP workspace binding (mandatory)
Before any ark_* MCP tool, call ark_identity with project.expectedRoot set to the exact
workspace root. Continue only when binding.status === "matched" and authoritative === true;
retain projectIdentity.projectId, then pass both expectedRoot and expectedProjectId under
project on every later MCP call. If identity is missing, mismatched, unverified, or the root is
uncertain, do not consume MCP analysis: use the workspace-local CLI and report that MCP
restart/retargeting is required. ark://manifest never satisfies this preflight.
Atomic multi-file work uses ark_prepare_change with the same matched project envelope.
Dual plane — layers + extras (mandatory, except /ark-runtime)
Label findings [Layer] vs [ArkRules] vs [ArkRun] vs [ArkOrder]. Absence of extras is valid and silent. Extra skip clusters (ARKRUN_* / ARKORDER_*) are /ark-autopilot. Never invent mechanical-safe kinds. Do not invent /ark-run or /ark-order. Skills never enforce.
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.
- 2d ago First seen · 86 lines · 22 tokens per session scan A c8a0a7589198
ark-loop is a skill published in the GitHub repository pedroknigge/arkgate (1 stars, last pushed 2d ago), licensed MIT. It adds 22 tokens to every session and 1,099 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.
code-review-architecture
Architecture-focused code review covering hexagonal boundary violations, DDD anti-patterns, CQRS misuse, and microservices coupling issues. Applied in addition to the language-specific review skill when architecture markers are detected. Invoked when reviewing hexagonal architectures, DDD patterns, or microservices…
tsq-ddd
Domain-Driven Design 전략적 설계 가이드라인. Bounded Context, Ubiquitous Language, Context Map. Use when: 도메인 모델링, Bounded Context 설계, 유비쿼터스 언어 정의, 컨텍스트 맵 작성 시.
ddd-aggregate-modeling
Guides aggregate modeling decisions for DDD services: aggregate root boundaries, child entity vs value object decisions, invariants, snapshot/primitives mapping, specification/criteria usage, and repository contract design. Invoked when the user asks to model or refactor domain entities and aggregate behavior.
AWS WAF Pillar Review
AWS Well-Architected Framework pillar checklists — operational excellence, security, reliability, performance efficiency, cost optimization, sustainability — for reviewing AWS infrastructure changes.
Azure WAF Pillar Review
Azure-specific Well-Architected Framework pillar checklists — reliability, security, cost optimization, operational excellence, performance efficiency — for reviewing Azure infrastructure changes.