The Gate

Source tier standard used for this record: T1 UN bodies · T2 established HR monitors · T3 official government/IGO records · T4 first-hand reputable news · T4-adj indirect access · secondary corroboration, commercial (not citable for fact) · not used.

Four bodies of evidence are tracked and never merged, summed, or netted against one another, even where their calendar windows overlap: the protest crackdown (December 28, 2025 to February 27, 2026), War Round 1 (February 28-April 7, 2026), War Round 2 (July 6 to 24, 2026), and post-MOU repression (June 1-August 10, 2026). A fifth window, the Islamabad MOU interim period (June 17-July 7, 2026), is carried as context only. The rule that keeps the boundary meaningful: a repression act that merely cites the war as pretext, such as an execution on a war-related espionage charge, stays in the repression record, because it is an act of the judiciary, not a war casualty. See The Record for the figures each lane actually carries.

This tracker publishes only material that has passed its evidentiary gate. A watch cluster, evidence row, or repression entry is published when it carries a Tier 1 to 3 source, or two independent Tier 4 outlets, or has been individually re-verified against a primary source by the project lead and dated as such. Material that has not yet cleared that bar is not published.

Where this tracker relies on a single accepted-tier monitor for a claim, that claim is published and it is not marked with a verdict badge or bracket flag: it is kept and rendered as an attributed statement in ordinary prose; “Hengaw reported…”, or “Iran International, citing HRANA, reported that…; no second outlet has corroborated the report”; so the reader always knows exactly who is making the claim and whether anyone else has confirmed it. Source attribution itself is never removed: a citation is not verification language, and every fact on this site keeps its source link regardless of how many outlets have confirmed it. This standard applies to every page.

No two monitors count the same way, and this project preserves that variance rather than manufacturing false precision. Figures flagged is_minimum in the underlying data are floors, not estimates of a true total, and are never averaged with other monitors’ floors to produce a single number. The clearest example is the crackdown death toll, which this site always presents as a range across methodologies rather than a single figure: the government’s official figure of 3,038 (UN Fact-Finding Mission, A/HRC/63/61, para. 31; its earlier January 21 figure was 3,117, per France24), HRANA’s minimum-verifiable 7,007, the UN Special Rapporteur’s medical-source estimate of at least 5,000 (up to ~20,000), and local/leaked estimates as high as ~36,500. The same doctrine applies to the post-MOU execution counts: June is reported as 101 to 127 and July as 66 to 71, by source, because Hengaw, HRANA, and IHRNGO use different methodologies and different cutoff dates. See Publication Standard above for how single-source figures and not-yet-corroborated items are handled.

A party appears in the register if any one of three things is true: it carries a designated, partial, or host-ban status from any of the five tracked jurisdictions, at any date (a pre-window designation is still public record); it appears on the Do Not Touch screen; or it documents a state official or perpetrator role on Tier 1 to 3 evidence.

Narrative text (the archived campaign report, watch-cluster summaries, dossier paragraphs, network corridor summaries and the evidence lines on a profile) renders as its source wrote it. A name that appears inside such a sentence (“linked to X”, “contracts with X”) is part of the sourced description of someone else’s conduct, not a listing.

This record combines three sets of research into one deduplicated register, with the origin of each entry tracked:

• Primary records. Designation registers, entity investigations, and casualty and watch briefings.
• Profile registry. The 743-profile registry that anchors most profile ids.
• Verified campaign research. The verified June 1 to August 10, 2026 research behind the public campaign report.

Where the sources compiled here conflict: primary records take precedence, then verified campaign research, then the profile registry.

This tracker will contain errors at this scale, and it is built to be corrected in the open. To report one, use the link in the footer of any page or write to info@meforum.org. Please cite the page and, where relevant, the profile id shown in the page URL.