This policy sets the standard for editorial work published by AgentsCalling. It applies to news, explainers, analysis, comparisons and product coverage.

Scope and story selection

We cover AI agents through Launches, Business, Developer and Analysis. Stories must establish a direct connection to autonomous or tool-using systems, agentic workflows, voice agents, multi-agent systems or the infrastructure and rules around them.

We select stories for their significance and practical relevance. Publishing frequency is not a reason to create filler or repeat a press release without additional value.

Sources and verification

We seek primary evidence such as official announcements, documentation, changelogs, research papers, regulatory notices and direct responses. We credit journalists and publications whose original reporting led to a story.

We check names, dates, figures, product access and important qualifications. We distinguish the date of an event from the date it was reported. Older developments must not be presented as new launches.

Company statements and research results are attributed to their sources. A press release is not independent verification of performance. Where evidence is incomplete, conflicting, paywalled or unavailable, we explain the limitation and avoid filling gaps with assumptions.

What we add

Each substantive article should resolve a useful question beyond the announcement: who can use the product, what permissions it requires, what changed in its documentation, how pricing works, what a benchmark actually measures or what remains unknown.

Analysis is identified as analysis. Predictions and implications must be distinguished from observed outcomes.

Hands-on testing

We use terms such as “we tested” only when the described test was actually performed. Test-based articles should explain the setup, relevant versions and dates, tasks, limitations and any access provided by a vendor. A demo or benchmark supplied by a company must not be represented as our own test.

Use of AI

AI tools may assist with discovery, research organisation, summarisation, drafting, editing and illustrations. The responsible editor remains accountable for the published content. Generated facts, quotations, sources and test results are not acceptable substitutes for evidence.

AI-assisted work requires source checking and editorial review. Generated or illustrative visuals must not imply they are documentary photographs, genuine screenshots or observed events. Where AI materially shapes a story or visual, we disclose its role in context.

Commercial relationships and conflicts

Sponsored material must carry a clear label. Affiliate links, paid access, free products or other material relationships relevant to an article must be disclosed. Payment does not buy a favourable editorial verdict.

Our founder’s involvement with Techlithic, Sendwo and HowToBuySaaS is described on our About page. Coverage of these businesses or related commercial interests must disclose the connection within the article. We do not describe such coverage as free from conflicts.

Fairness, bylines and accountability

Articles should identify the author or responsible editorial desk. We aim to seek relevant responses before publishing substantial allegations and to describe unanswered requests accurately. We do not invent interviews, quotations, expertise or newsroom staff.

Sample content is labelled as such. Material errors are handled under our Corrections Policy. Send concerns through Contact.

AgentsCalling. News. Context. Less noise.