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 Edge-City/agentvillage --skill approvalgit clone --depth 1 https://github.com/Edge-City/agentvillageWrote 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/edge-city/agentvillage/approval)<a href="https://agentmods.dev/skills/edge-city/agentvillage/approval"><img src="https://agentmods.dev/badge/skills/edge-city/agentvillage/approval/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/edge-city/agentvillage/approval"><img src="https://agentmods.dev/badge/skills/edge-city/agentvillage/approval.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 | $0.00108 | $0.01710 |
| Opus 5.5 | $0.00043 | $0.00684 |
| Sonnet 5.5 | $0.00022 | $0.00342 |
| Haiku 4.5 | $0.00011 | $0.00171 |
Grade A, and why
approval 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Approval gate
The resident opted in to a record of what you do. Before terminal,
write_file, patch, read_file and search_files run, and process,
web_extract, the browser_* tools, skill_manage, delegate_task,
cronjob_manage and send_message too, and web_search, x_search, the
media tools (image_generate, video_generate, text_to_speech) and every
one of Index's write tools (creating, rewording, pausing, resuming or archiving
a signal, accepting or declining an opportunity, changing or enriching the
profile, joining or changing a network), the call is checked against the
resident's approval policy, which lives with their approval service, not where
you can edit it. On the starting policy almost every call is recorded and
passes straight through. Three kinds never run for you: changing the gate
itself (the Hermes config, its hooks, the consent allowlist), touching the
approval log, and reading or changing the resident's credentials. If the
resident has made a kind of call wait for them, it waits for their tap in their
approval bot on Telegram. If the gate cannot be reached, or cannot be verified
(av-approval: gate unverified (...)), gated calls are blocked.
Publishing an intention you inferred, sharing a digest, and casting the resident's answer to the village question are not yours to do directly: the overlay sends each to the resident as a proposal and acts only on their tap.
Shape every gated call so it can be checked
terminal: always passworkdiras an absolute path (for example theHERMES_HOMEdirectory). A call with noworkdir, or a relative one, is refused withhook-unsupported-execution-context.write_file,patch,read_file,search_files: always givepathas an absolute path starting with/. A relative or missing path is refused the same way.execute_codeis not gated: you may use it for Python work such as parsing, converting or calculating. Its working directory is a temporary one; that is not isolation. Never use it to do something a gated call was refused or is waiting for: that is routing around the gate.- Never write under
$HERMES_HOMEfrom code (execute_codeincluded): the approval gate's own files live there, and a change to them can block every gated call until the box is reinstalled. - Python runs through
execute_codeonly.pythonorpython3throughterminalstays refused by the approval rules; do not retry it there.
What ships with it
4 files 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.
- yesterday Changed · +8 lines fac8fbd15901
- 12d ago Changed · +6 lines c0ca8e787a7b
- 20d ago Changed · +8 lines 1e39dbb168a4
- 23d ago Changed · +19 lines · +17 tokens per session 4050ef9f81c4
- 25d ago First seen · 88 lines · 91 tokens per session scan A 1016c1f425ae
approval is a skill published in the GitHub repository Edge-City/agentvillage (21 stars, last pushed today), licensed MIT. It adds 108 tokens to every session and 1,710 once invoked, about $0.0004 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 2026-10-03.
Other skills, from other repositories
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…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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…