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 rcarmo/piclaw-addons --skill remote-peergit clone --depth 1 https://github.com/rcarmo/piclaw-addonsWrote 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/rcarmo/piclaw-addons/remote-peer)<a href="https://agentmods.dev/skills/rcarmo/piclaw-addons/remote-peer"><img src="https://agentmods.dev/badge/skills/rcarmo/piclaw-addons/remote-peer/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/rcarmo/piclaw-addons/remote-peer"><img src="https://agentmods.dev/badge/skills/rcarmo/piclaw-addons/remote-peer.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 | $0.00023 | $0.00720 |
| Opus 5.5 | $0.00009 | $0.00288 |
| Sonnet 5.5 | $0.00005 | $0.00144 |
| Haiku 4.5 | $0.00002 | $0.00072 |
Grade A, and why
remote-peer 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 16d 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.
What it actually says
Iroh Remote Peer
Operators pair clients in Settings by pasting PCL1-… client IDs, comparing identity with the owner and explicitly approving requests. Internet address lookup is a separate opt-in for bare-ID dialing. Advanced endpoint tickets work without lookup. mDNS is off by default, settable in Settings, and never auto-pairs.
Agents use chat({action:"directory"}) and only returned addresses/modes, such as lab!inbox and lab!@research. Send workspace files or media_ids; do not encode binary in message text. Use stable idempotency_key values for uncertain retries. Reply to supplied peer!reply.… addresses unchanged.
Operator remote_peer actions: status, identity, ticket, pair, accept, deny, revoke, forget, alias, policy, remote_permissions, advertise, unadvertise, session_targets, session_advertise, session_unadvertise, ping, retry, work_send, work_review. Pair uses client_id with optional alias and ticket. Acceptance/revocation/removal require full client-ID confirmation. Wider incoming permissions require ALLOW REMOTE ACCESS.
Settings exposes per-peer incoming drafts with Apply/Cancel/Revert and a separate read-only outgoing advertisement. remote_permissions explicitly refreshes one paired peer's remote roster; it does not grant permissions. To send files to a peer, the receiver must enable incoming files for this instance, then the sender refreshes its directory. File-only edits preserve scope, named agents and modes. No networking is enabled automatically.
Session targets are operator opt-in. Advertise the local agent first, then use Advertised sessions to choose an exact local target, advertised agent alias, public session alias and modes. Idle sessions are valid. Each peer needs an explicit session grant in the existing incoming draft and typed ALLOW REMOTE ACCESS consent. Neither named-agent selection nor all-advertised grants sessions implicitly. File-only edits preserve session grants.
Use peer!@agent/session only when returned by the directory, with its returned modes and status. Outgoing Settings displays read-only session grants. Stale incarnations reject delivery without named-agent fallback; older hosts without incarnation-bound APIs report unavailable/unreachable targets. Hiding an advertisement clears its grants. Every advertisement/re-advertisement creates a fresh revision and requires fresh grants, even for the same lifetime. Explicit policy.sessions grants must include the selected advertisement's revision alongside agent, session_alias and modes; never invent or reuse a stale revision. Advertisement changes invalidate typed consent even with no saved grants: use Revert, reselect the current session and confirm afresh. Retyping or toggling a stale draft cannot unblock Apply. Session management requires request-scoped local authority; never invent a chat JID or incarnation.
Pairing grants no direct remote tool execution. Work proposals and execute-labelled requests both require review and a locally supplied result.
This implementation uses only Iroh peer transport and fresh state. HTTP URLs, the old /pair command, old tool actions and old peer databases are unsupported. Do not migrate or delete legacy installed data. Do not enable networking, public lookup or mDNS without operator direction.
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.
- 16d ago Changed · +4 lines 211be9ed5d57
- 1mo ago Changed · +2 lines 64e5b5c8e78d
- 1mo ago Changed · -47 lines · -5 tokens per session e7f36fb89386
- 2mo ago First seen · 65 lines · 28 tokens per session scan A 4963eaf8a056
remote-peer is a skill published in the GitHub repository rcarmo/piclaw-addons (23 stars, last pushed yesterday), licensed MIT. It adds 23 tokens to every session and 720 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
pi
Use when you need a minimal, extensible terminal AI coding agent harness in TypeScript — npm @earendil-works/pi-coding-agent 1.4.0, extensions/skills/prompt templates, multi-provider (OpenAI/Anthropic/Google) routing, container isolation (Docker/Gondolin/OpenShell); also for pi: command not found, Node >=22.19.0…
OpenShell + Pi: build an isolated coding-agent sandbox
Use when the user wants to install, configure, migrate, harden, inspect, or troubleshoot a Pi coding agent running in NVIDIA OpenShell, especially with rootless Podman, deny-by-default networking, hidden credentials, persistent state, or A2A communication. Do not use for ordinary containerization that does not involve…
ensembl-database
Query Ensembl genome database REST API for 250+ species. Gene lookups, sequence retrieval, variant analysis, comparative genomics, orthologs, VEP predictions, for genomic research.
skill-news-intelligence
A daily industry-news research workflow that gathers in-depth articles and updates from Chinese technology, business, creator, and AI publications. It filters the results by a creator’s field and turns them into a brief with possible content topics.
ai-music
A workflow that turns a topic or script into a finished short video by combining writing, images or video clips, spoken narration, subtitles, background music, and final assembly.
sidewinder-rattlesnake
Adversary-emulation profile for SideWinder (G0121 / Rattlesnake / T-APT-04 / Razor Tiger), India's suspected state-sponsored cyber-espionage actor.