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/pushary/pushary-skill/hermesnpx skills add Pushary/pushary-skill --skill hermesgit clone --depth 1 https://github.com/Pushary/pushary-skillWhat 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 | $0.00164 | $0.02187 |
| Opus 5 | $0.00082 | $0.01094 |
| Sonnet 5 | $0.00033 | $0.00437 |
| Haiku 4.5 | $0.00016 | $0.00219 |
Grade C, and why
pushary-hermes scanned grade C with 1 finding 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
"question": "Allow: rm -rf /tmp/build-artifacts/*", How it starts
The opening of the file, as written. The whole thing — 204 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pushary - Push Notifications for Hermes Agent
Pushary adds web push notifications as a delivery channel for Hermes. Use it when the user is not actively monitoring a chat platform, or when you need to reach them on their phone's lock screen for a time-sensitive decision.
Ask in as Few Interruptions as Possible
Every question costs the user their attention wherever they are. Before a run of more than a step or two, work out where you will need a human and fold those points together: one select carrying the real options beats three confirms in a row, ask once at a boundary rather than once per instance, and never ask what you can determine yourself from the task or from a tool call you can make.
When to Use Pushary vs Hermes Platforms
Use Pushary when:
- The user has no active chat session (Telegram, Discord, etc.)
- You need to reach the user's phone lock screen for a quick decision
- A background task finishes and the user may have walked away
- Permission escalation - a dangerous command needs approval
- Another skill's workflow says to confirm with the user, and no chat session is active to confirm in
- The user explicitly asked for push notifications
Use the active Hermes platform when:
- The user is currently in a Telegram/Discord/Slack conversation with you
- The question is part of an ongoing dialog
- The user prefers responses in their current platform
Use both when:
- A critical error occurs - notify via push AND the active platform
- A long-running task completes - push ensures they see it even if they closed the chat
Setup
npx @pushary/agent-hooks@latest setup --agents hermes
export PUSHARY_API_KEY="pk_xxx.sk_xxx"
That installs hermes-plugin-pushary into the interpreter Hermes runs in, enables it, and registers the tools natively. No MCP server config is needed. Sign up at https://pushary.com/sign-up?from=hermes to get your API key.
Approvals Go to the Phone
The plugin registers pushary as a Hermes approval transport, so the dangerous-command approvals Hermes already asks for are answered from the lock screen with the same four choices the terminal offers: allow once, allow for this session, always allow, deny. Hermes owns the timeout (300 seconds by default) and remembers session and always decisions exactly as it would have.
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 First seen · 204 lines · 164 tokens per session scan C effeca2a9b6f
pushary-hermes is a skill published in the GitHub repository Pushary/pushary-skill (0 stars, last pushed 4d ago), licensed MIT. It adds 164 tokens to every session and 2,187 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
social.reply
Reply to a recent direct or group message without inventing unrelated movement.
kukuroo-notify
Push a notification to the user's own phone through their Kukuroo Worker. Use when they ask to be notified, pinged, buzzed, or sent a status update on their phone; name a time they want to be told at; say they are stepping away from the monitor or will be afk; or mention Kukuroo. Also for diagnosing a Kukuroo send…
surface
Surface-native display for AI agents, driven by the surface CLI. Use when the user says "surface this", "show me X", or "put it on my display/screen"; wants a live interactive UI, chart, or tool they act on; needs a question answerable from any device; or asks you to react to what they click — even while you're…
meeting-copilot
A two-machine live meeting copilot. A Windows laptop streams two audio channels (your mic, and a loopback of the other side's voice) to a parent machine, which transcribes them, drives a teleprompter you read from, watches every incoming utterance against a ledger of pre-agreed facts, and answers off-script questions.…
batch
Batch job-application filler. Given several job URLs (or a urls.txt file), run the assisted-apply flow for each in its own browser tab — fit-gate, fill-or-flag, stop before Submit — skipping stale or low-fit postings, then present one combined review table so the human can approve each tab one by one. Use for /batch…
board
Application tracker board. Read /.dear-hiring-manager/applications.md and render a status board (grouped by blocked/skipped/filled/submitted/rejected/interview/offer with counts and a response-rate summary), and update an application's status + date when the user reports progress (e.g. "mark Transcend as interview").…