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 shawnla90/gtm-coding-agent --skill instantly-hubspot-repliesgit clone --depth 1 https://github.com/shawnla90/gtm-coding-agentWrote 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/shawnla90/gtm-coding-agent/instantly-hubspot-replies)<a href="https://agentmods.dev/skills/shawnla90/gtm-coding-agent/instantly-hubspot-replies"><img src="https://agentmods.dev/badge/skills/shawnla90/gtm-coding-agent/instantly-hubspot-replies/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/shawnla90/gtm-coding-agent/instantly-hubspot-replies"><img src="https://agentmods.dev/badge/skills/shawnla90/gtm-coding-agent/instantly-hubspot-replies.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 | $0.00161 | $0.05430 |
| Opus 5.5 | $0.00064 | $0.02172 |
| Sonnet 5.5 | $0.00032 | $0.01086 |
| Haiku 4.5 | $0.00016 | $0.00543 |
Grade A, and why
instantly-hubspot-replies 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 today.
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 — 276 lines — stays where its author put it; the contents beside it link to each section on GitHub.
instantly-hubspot-replies
I log Instantly replies onto the HubSpot contact as real email engagements, one thread per lead. A positive reply opens an unowned deal at MQL so the team can route it. The card replies through Instantly from the same cold sending mailbox. HubSpot's own Reply button sends from the rep's inbox and breaks that thread.
The decisions are in reply_loop.py. Tests are tests/test_reply_loop.py. They use no network. Chapters 26, 27, and 28 are the narrative. This skill is the build, then the gotchas I hit after it was live. Read the gotchas before you write a record. A retired mailbox is dead-address-refind and Chapter 29.
Names, portal ids, campaign ids, and domains below are placeholders. The client is "the client". The campaign is "a conference campaign". The guard string is CAMPAIGN_NAME_GUARD. The allowlist domain is client-domain.com.
Inputs
HUBSPOT_ACCESS_TOKEN(private app service key),HUBSPOT_PORTAL_ID,HUBSPOT_ALLOW_<CLIENT>_PORTAL(this starter usesHUBSPOT_ALLOW_CLIENT_PORTAL)INSTANTLY_API_KEY,INSTANTLY_CAMPAIGN_IDS,INSTANTLY_CAMPAIGN_DEAL_PREFIXESINSTANTLY_TEST_CAMPAIGN_GUARD(placeholder valueCAMPAIGN_NAME_GUARD). Honored only while test mode is on. See gotcha 3.HUBSPOT_MQL_PIPELINE_ID,HUBSPOT_MQL_STAGE_ID(ids, not the label MQL)INSTANTLY_WEBHOOK_SECRET,CRON_SECRET,HUBSPOT_APP_SECRET(from the app Auth tab, never fromvercel env pull)HUBSPOT_REPLY_ALLOWLIST(@client-domain.com, or full emails of the client's logged-in HubSpot users)INSTANTLY_REPLY_TEST_MODE=trueuntil the first hand-checked reply has been applied
How to run
python3 -m unittest tests/test_reply_loop.py
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.
- today First seen · 276 lines · 161 tokens per session scan A fa73493961ce
instantly-hubspot-replies is a skill published in the GitHub repository shawnla90/gtm-coding-agent (170 stars, last pushed today), licensed MIT. It adds 161 tokens to every session and 5,430 once invoked, about $0.0006 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 2026-11-13.
instantly-hubspot-replies compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| instantly-hubspot-repliesshawnla90/gtm-coding-agent | Skill | 170 | today | MIT | A | 161 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 135k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 100k | today | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 80k | 6d ago | MIT | A | 200 | Claude Code, Codex |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| chat-pet-sprite-creationmicrosoft/vscode | Skill | 195k | today | MIT | A | 53 | Claude Code, Codex |