<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>AI agents for saas · Agentic Feats</title><description>Real examples of AI agents in saas.</description><link>https://agenticfeats.com/</link><language>en</language><item><title>SaaS accounting automation</title><link>https://agenticfeats.com/c/saas-accounting-automation-hermes-agent/</link><guid isPermaLink="true">https://agenticfeats.com/c/saas-accounting-automation-hermes-agent/</guid><description>Hermes Agent, connected to Zoho Books, Gmail, Aspire Bank, Stripe and Telegram, ran bookkeeping for the author&apos;s Singapore SaaS company. Result: Setup took about 6 to 8 hours of active work and cut routine month-end bookkeeping by an estimated 85 to 90 percent. Evidence: Numbers. Source: https://satvasolutions.com/blog/how-we-automated-accounting-ai-agent</description><pubDate>Tue, 13 Oct 2026 15:29:03 GMT</pubDate><category>Hermes Agent</category><category>Finance and Admin</category></item><item><title>Backlink outreach for Promptslove</title><link>https://agenticfeats.com/c/backlink-outreach-for-promptslove/</link><guid isPermaLink="true">https://agenticfeats.com/c/backlink-outreach-for-promptslove/</guid><description>The author&apos;s ChatGPT Dot, Jamie, researched publishers, drafted pitches, sent outreach and form submissions, and tracked results for Promptslove&apos;s digital PR campaign. Result: Jamie reported 13 outreach actions and zero live links early on; the author later verified the first live nofollow directory backlink. Evidence: Shown. Source: https://promptslove.com/blog/openai-dots-guide</description><pubDate>Tue, 13 Oct 2026 15:29:03 GMT</pubDate><category>Other</category><category>Marketing and Content</category></item><item><title>Shared team Hermes agent</title><link>https://agenticfeats.com/c/shared-team-hermes-agent/</link><guid isPermaLink="true">https://agenticfeats.com/c/shared-team-hermes-agent/</guid><description>Hermes Agent spread from the author&apos;s own use to their dev, co-founder, and a shared team agent named Robert. Result: Robert pulls context from calls, helps with approvals, and creates tasks in Multica, backed by shared memory. The non-technical co-founder reportedly gets more done. Evidence: Claim. Source: https://x.com/niklas11891/status/2105541804267274335</description><pubDate>Tue, 13 Oct 2026 15:29:03 GMT</pubDate><category>Hermes Agent</category><category>Business</category></item><item><title>Claude in Chrome PM review</title><link>https://agenticfeats.com/c/claude-in-chrome-pm-review/</link><guid isPermaLink="true">https://agenticfeats.com/c/claude-in-chrome-pm-review/</guid><description>Claude in Chrome was tested on help center docs, Google Ads campaign setup, and a Looker dashboard by a non-engineer product manager. Result: Docs rated 2 of 5, Google Ads setup 4 of 5, Looker dashboard 5 of 5; best docs run took about 2 hours. Evidence: Shown. Source: https://rimo.app/en/blogs/claude-in-chrome-review1_en-US</description><pubDate>Tue, 13 Oct 2026 15:29:03 GMT</pubDate><category>Claude in Chrome</category><category>Business</category></item><item><title>Support automation skill builder</title><link>https://agenticfeats.com/c/support-automation-skill-builder-claude-code/</link><guid isPermaLink="true">https://agenticfeats.com/c/support-automation-skill-builder-claude-code/</guid><description>A Claude Code skill that guides a team through building custom support automation across ticketing, billing, and database platforms, with triage and templated replies. Result: The Lex support automation built this way has 15 scripts and 6 templates, and handles about 80% of support tasks with yes/no approvals. Evidence: Shown. Source: https://github.com/nbashaw/claude-cs</description><pubDate>Tue, 13 Oct 2026 15:29:03 GMT</pubDate><category>Claude Code</category><category>Business</category></item></channel></rss>