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 SII-Holos/synergy --skill change-execution-boundariesgit clone --depth 1 https://github.com/SII-Holos/synergyWrote 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/sii-holos/synergy/change-execution-boundaries)<a href="https://agentmods.dev/skills/sii-holos/synergy/change-execution-boundaries"><img src="https://agentmods.dev/badge/skills/sii-holos/synergy/change-execution-boundaries/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/sii-holos/synergy/change-execution-boundaries"><img src="https://agentmods.dev/badge/skills/sii-holos/synergy/change-execution-boundaries.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00070 | $0.01180 |
| Opus 5 | $0.00035 | $0.00590 |
| Sonnet 5 | $0.00014 | $0.00236 |
| Haiku 4.5 | $0.00007 | $0.00118 |
Grade A, and why
change-execution-boundaries 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 — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Change Execution Boundaries
Trace the Whole Decision
- Read Execution boundaries and
packages/product-runtime/AGENTS.md. - Start at
session/tool-resolver.ts, then trace the operation through capability classification, the enforcement gate, profile compilation, saved/session permission layers, SmartAllow, approval side effects, sandbox policy, and the tool implementation. - Inspect
packages/util/src/capability.tsfor the shared capability catalog and public severity/category metadata. Keep classification independent from profile policy: classifiers describe what an operation can do; profiles decide allow, ask, or deny. - Check built-in tools, plugin and MCP envelopes, worktree/main-checkout reclassification, sensitive paths, remote execution, and platform fallback before assuming one call site owns the boundary.
Preserve the Security Model
- Route every executable path through the centralized gate. Do not move authorization into one tool or treat visibility, authorization, and sandboxing as the same decision.
- Keep model presentation and execution separate.
ToolCatalogsent to an Agent worker contains only serializable schemas; executable callbacks and permission state remain in the Control Plane and enterToolScheduleronly after the Agent stream is disposed. - Keep production capability analysis in the bounded Policy worker pool. The worker receives immutable classification context and returns capabilities; profile decisions, approvals, audit state, sandbox accumulation, and execution remain in the Control Plane. Await a bounded worker handshake before admission, and turn worker infrastructure failure into a direct conservative denial rather than an approval request.
- Make classification termination explicit. A parser or classifier must share one time/depth/active-input budget and return a finite conservative capability on no progress, repeated input, or excessive depth. It must never restart its public top-level entry point to handle those conditions.
- Keep
guardedinteractive,autonomousnon-prompting, andfull_accesspermission-silent without suppressing ordinary runtime failures. - Mark hard boundaries non-bypassable at the capability definition or classification source. SmartAllow and preauthorization must not override hard denials or receive raw secrets.
- Keep the active workspace as the default write/execute boundary. Preserve original-checkout and sibling-worktree protection, trusted-root containment, protected metadata, and credential-path rules.
- Treat sandboxing as post-authorization containment. Preserve filesystem roots, network mode, approved external roots, shell-bypass semantics, and the configured
deny/warn/allowfallback. - Keep permission policy and the
SandboxHostcontract inpackages/harness; OS backends, PTYs, filesystem watchers and helper sources belong topackages/runtime-local. Register the host and native startup contribution in local composition, including agent-worker bootstrap. Missing host registration must fail closed. Change platform helpers and TypeScript policy together. Do not claim parity without the relevant macOS Seatbelt, Linux helper/Bubblewrap, Windows helper, or WSL evidence.
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.
- 2d ago Changed · +2 lines · -1 tokens per session 74489f995f99
- 10d ago First seen · 42 lines · 71 tokens per session scan A 94b35991c079
change-execution-boundaries is a skill published in the GitHub repository SII-Holos/synergy (490 stars, last pushed today), licensed MIT. It adds 70 tokens to every session and 1,180 once invoked, about $0.0003 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 2026-08-30.
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…