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/osouthgate/agent-plus/agent-plus-installernpx skills add osouthgate/agent-plus --skill agent-plus-installergit clone --depth 1 https://github.com/osouthgate/agent-plusWrote 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/osouthgate/agent-plus/agent-plus-installer)<a href="https://agentmods.dev/skills/osouthgate/agent-plus/agent-plus-installer"><img src="https://agentmods.dev/badge/skills/osouthgate/agent-plus/agent-plus-installer.svg" alt="Measured on agentmods" 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 | $0.00110 | $0.01719 |
| Opus 5 | $0.00055 | $0.00860 |
| Sonnet 5 | $0.00022 | $0.00344 |
| Haiku 4.5 | $0.00011 | $0.00172 |
Grade C, and why
agent-plus-installer scanned grade C with 2 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 4d 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsSL https://raw.githubusercontent.com/osouthgate/agent-plus/main/install.sh | sh -s -- --unattended Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
(e.g., curl + jq + curl against the same provider) — agent-plus probably How it starts
The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agent-plus-installer
A pure-markdown skill that gives Claude Code precise triggers for when to offer to install agent-plus on a user's behalf, plus the safety rails for how. No bin, no Python — the skill is the documentation; the agent is the runtime.
The capability already exists (shipped in agent-plus 0.12.0): install.sh --unattended chains into agent-plus-meta init --non-interactive --auto
which emits a frozen JSON envelope. This skill ships the trigger doctrine
so Claude's skill router fires it on the right cues — not three times a
session.
Killer command
curl -fsSL https://raw.githubusercontent.com/osouthgate/agent-plus/main/install.sh | sh -s -- --unattended
That single line is the install. The agent types one thing, not two. The internal pipeline (install.sh -> agent-plus-meta init --non-interactive --auto) is documented under Architecture below — but the agent should never compose it by hand. Following the killer command verbatim succeeds; reconstructing the pipeline invents flags that don't exist.
Do NOT use this for
- User already has agent-plus. Probe with
command -v agent-plus-metabefore offering. If it resolves, stop — they're installed. (Useagent-plus-meta --versionto surface what version they're on if relevant, but don't offer install.) - User declined this session. Honor the session-scope decline flag (Safety rule #3). Re-offering in the same session is the spam-pop-up failure mode.
- False-positive trigger. If the user mentioned a service name in passing but isn't actually doing work agent-plus would help with, do not offer. "I deployed to Vercel yesterday" is not a trigger. "I keep typing the same five Vercel CLI commands and it's slow" is.
- Sandbox or ephemeral CI environment. If
$CI,$GITHUB_ACTIONS,$CODESPACES, or similar env vars are set, or the working directory is under/tmp, do not offer — the install will not persist and will waste user time.
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.
- 4d ago First seen · 149 lines · 110 tokens per session scan C 56779bb3affc
agent-plus-installer is a skill published in the GitHub repository osouthgate/agent-plus (4 stars, last pushed 1mo ago), licensed MIT. It adds 110 tokens to every session and 1,719 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
notify-user
Telegram comms playbook for Antigravity — when and how to reply, ack long work, present choices, and delegate to other agents. Use whenever you're paired to a Telegram chat via the telegram-agy MCP server and need to talk back to the user. The chat is your only signal channel — terminal output never reaches the user.
rag-your-code
Index and search the current codebase with local, explainable RAG records.
slack-cli
Complete Slack Web API CLI — 201 methods across 29 command groups. Every public endpoint. Agent-friendly JSON output. Auto-generated from the official OpenAPI spec + @slack/web-api SDK. Use when: sending messages, reading channels, adding reactions, managing files, searching, pins, bookmarks, reminders, user groups…
slack-messages
Send Slack messages to channels or users via the elnora-slack CLI. Resolves channel/user IDs from the cached workspace reference, formats messages, handles threads and DMs, and enforces a draft-and-approve gate before anything is sent. TRIGGERS: "send slack", "slack message", "post on slack", "message on slack", "tell…
compounded-author
Use this skill when you have just completed a non-trivial task (more than 3 tool calls, novel approach, recovery from a dead end) and are considering whether to save the procedure as a reusable skill via compounded; OR in RULE MODE when the user corrected your approach and the correction encodes a generalizable…
compounded-verifier
Use this skill ONLY when invoked as the skill-verifier subagent by compounded's Stop hook. This is the rubric for deciding whether a proposed skill should graduate to .verified or move to .rejected/ based on its replayability against a just-completed task. Do not invoke this skill in normal sessions.