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 melodic-software/claude-code-plugins --skill reachgit clone --depth 1 https://github.com/melodic-software/claude-code-pluginsWrote 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/melodic-software/claude-code-plugins/reach)<a href="https://agentmods.dev/skills/melodic-software/claude-code-plugins/reach"><img src="https://agentmods.dev/badge/skills/melodic-software/claude-code-plugins/reach/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/melodic-software/claude-code-plugins/reach"><img src="https://agentmods.dev/badge/skills/melodic-software/claude-code-plugins/reach.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 | $0.00226 | $0.02501 |
| Opus 5 | $0.00113 | $0.01251 |
| Sonnet 5 | $0.00045 | $0.00500 |
| Haiku 4.5 | $0.00023 | $0.00250 |
Grade A, and why
reach 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 — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
Getting an agent turn to happen on a machine that is not this one. The obvious route, asking the peer tools to find the other machine's session, does not work here and cannot be made to: the accounts are split on purpose. This carries the route that does, plus the reasons the shortcuts fail, so neither gets rediscovered by trial.
Read ~/.config/fleet/FLEET.md before composing anything. It is rendered per machine from the
fleet manifest and carries the real hosts, ports and accounts; the recipes below are the same
commands with placeholders where it has values.
Resolve the target first
Machines are addressed by hostname over the tailnet (MagicDNS), never by session name. A session name identifies a session, not a host, and nothing routes on it.
- Read
~/.config/fleet/FLEET.md. Done when you can name the target's MagicDNS host name and see its per-host table of reach entries, each carrying anid, transport, sshalias, port, account and shell. - Pick the reach entry by what the work needs, not by what is first in the table. Agent work goes
to
wsl-shell(port 2222, the distro account, bash); Windows PowerShell work goes towindows-shell(port 22,ssh-admin, pwsh). Done when you hold a concrete port and account for the command you are about to compose, both read from the table rather than assumed. - If the host is absent from that file it is not a fleet host. Done by saying so and stopping; the manifest is the only inventory, and a guessed name reaches nothing.
Verify the hop before you trust it
Cheap, and it separates "the host is unreachable" from "the agent over there is not signed in":
ssh -p 2222 <wsl-user>@<host> 'hostname && claude -p "echo ok" < /dev/null'
The hostname proves which machine answered. ok proves that machine's Claude is authenticated and
working. Do this before sending a long or expensive prompt.
One shot
ssh -p 2222 <wsl-user>@<host> 'claude -p "<prompt>" < /dev/null'
What ships with it
2 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 First seen · 170 lines · 226 tokens per session scan A 564eebea7515
reach is a skill published in the GitHub repository melodic-software/claude-code-plugins (16 stars, last pushed today), licensed MIT. It adds 226 tokens to every session and 2,501 once invoked, about $0.0011 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 2026-09-08.
Other skills, from other repositories
parallel-orchestrator
Manage parallel Claude Code workstreams using git worktrees. Use when: splitting large tasks across multiple workers, coordinating parallel development, monitoring worker progress, integrating completed work, analyzing work item documents (code reviews, issue lists). Triggers: parallel, orchestrator, worktrees…
parallel-worker
Execute focused implementation tasks in a parallel workflow. Use when: working on assigned files in a worktree, making checkpoint commits, signaling dependencies or blockers, completing orchestrator-assigned tasks. Triggers: worker, checkpoint, worktree, assigned scope, commit prefix, parallel task.
catch-expected-errors
For iteration with errors: catch exceptions during exploration, skip invalid cases, continue to next attempt.
count-combinations
For probability and counting: permutations, combinations, sample spaces, Monte Carlo simulation, brute-force enumeration, card/dice problems.
find-shortest-path
For pathfinding and search: shortest path, maze solving, game AI, route planning, graph traversal, BFS/DFS, Dijkstra, A problems.
generate-tree-structure
For tree/maze generation: spanning trees, random mazes, graph coverage. Uses frontier-based exploration with configurable traversal order.