Methodology: how the Islamic content here is checked

Where the text comes from

All 21 dhikr presets live in a single file in the repository. Each entry carries its Arabic, transliteration, English meaning and a direct source link: hadith entries to sunnah.com and Qur'anic entries to Quran.com.

The rule written at the top of that file is that nothing is sourced from memory, and that every reference in it was checked during research and again during verification. That's an editorial rule, not something code can enforce. The check you get is the source link, which sits on every entry: tap the info button next to the phrase and it opens the narration itself.

The post-salah program ships 9 sourced steps. It starts with istighfar and the salam formula, continues through the 33s and closing tahlil, then closes with the three Quls.

Ayat al-Kursi is available as a verified preset but is deliberately outside the guided sequence. Its post-salah narration is in a collection this site cannot link to directly, and a guided step has to carry a source a reader can open.

Qur'anic text is stored in the same reviewed data file after being compared against primary Qur'an sources. The app does not depend on a runtime Quran API, so the verified text continues to work offline.

• Astaghfirullah, 3 times. Sahih Muslim 591.
• The salam formula, once. Sahih Muslim 591.
• SubhanAllah 33, Alhamdulillah 33, Allahu Akbar 33. Sahih Muslim 597a.
• The closing tahlil, once, to complete the hundred. Sahih al-Bukhari 844.
• Al-Ikhlas, al-Falaq and an-Nas, once each. Sunan Abi Dawud 1523 and Sunan an-Nasa'i 1336.
• Ayat al-Kursi: available as a separate verified preset, but not a guided step because its post-salah narration is outside the collections this site can link to directly.

The counting rules

If the hadith gives no number, the preset ships with no target at all. The count field is null, the counter prints "no set count" where the target usually sits, and in the ring style the ring stays empty.

8 of the 21 presets are in that state. For those you count for as long as you want, because nothing in the text tells you when to stop.

Each preset opens on the count its source gives, or on no target when no count is narrated. Quick goals and the custom-target field are separate personal controls; choosing one changes your own goal, not what the source is claimed to narrate.

Where a familiar number is convention rather than narration, the count note says which is which. The two heavy words carries the line "The hadith gives no count. 100 is common practice, not a narrated number."

• The two heavy words. Sahih al-Bukhari 6682.
• La hawla wa la quwwata illa billah. Sahih al-Bukhari 6384.
• La ilaha illallah. Jami' at-Tirmidhi 3383.
• Durood Ibrahim. Sahih al-Bukhari 3370.
• Short salawat. Sahih Muslim 408.
• Hasbunallahu wa ni'mal-wakeel. Quran 3:173, Sahih al-Bukhari 4563.
• Talbiyah. Sahih al-Bukhari 1549. Recited continuously, so no fixed count.

How gradings are reported

15 of the 21 presets cite Sahih al-Bukhari or Sahih Muslim, where the collection is the grading, so no separate grading line appears on them.

1 preset carries its own grading line, and it's the one where the grades don't agree. La ilaha illallah cites Jami' at-Tirmidhi 3383, and the info sheet prints all 3: Tirmidhi's hasan gharib, al-Albani's hasan, and al-Hakim's sahih with al-Dhahabi agreeing.

The rule written at the top of the dhikr file is that every entry carries its grading, including when sunnah.com's printed grade differs from a scholar's. That Tirmidhi entry is the only shipped preset that exercises it, and all 3 grades stay on the page, including the 2 that fall short of sahih.

17 hadith-based presets link to sunnah.com, while the Qur'anic presets link to their surah or verse on Quran.com. Printing only the grade that flatters an entry would break the one thing that makes the citation useful.

Weak narrations

No preset in the app carries a weak grading today. Hadith-based entries cite their collections directly, and Qur'anic entries are identified by surah or verse.

The rule for anything added later: a weak narration ships with its grading printed on the entry, or it doesn't ship. The grading field that carries the Tirmidhi line is the same field a da'if grade would go in, so the place to put it already exists. Nothing has needed it yet, so read this as a stated policy with no shipped case behind it.

The same goes for popular narrations whose status is contested. Hasbunallahu wa ni'mal-wakeel carries a note that the 7-times morning and evening version is disputed, and that version isn't a preset.

Where scholars differ

Both positions get shown and neither is picked for you. The site names the scholar or the collection behind each position rather than ruling between them.

The Tirmidhi grading line is the clearest case in the app: 3 gradings, printed together, none of them settled.

The same discipline applies to narrations that often get merged into one. 3 of the count notes exist for that reason alone.

• Allahu Akbar. 33 is the Abu Hurayrah narration (Muslim 597a). The 34 form is a different narration, from Ka'b ibn 'Ujrah (Muslim 596a). The app ships 33 and prints the other one next to it.
• Astaghfirullah. 3 after the salam (Muslim 591). Separately, 100 a day (Muslim 2702) and "more than 70" (Bukhari 6307) are 2 different narrations with different wordings, not one hadith.
• The closing tahlil. Once after salah, to complete the hundred (Bukhari 844). As a daily dhikr it's 100 (Bukhari 6403, Muslim 2691). The note keeps the two apart.

Who checks this, and where that stops

Mufti Salim Qasmi builds and runs this site and reviews its Islamic content. He's a Hafiz of the Quran and an Aalim and Mufti from Darul Uloom Deoband, where he completed Dars-e-Nizami and the Ifta specialisation in 2019. The full record is on the about page.

The qualification belongs to the person, not to this website. Darul Uloom Deoband has not reviewed or endorsed this site, and nothing here should be read as its position.

This is a reference tool, not a fatwa service. It doesn't replace asking a scholar about your own situation, your madhhab or your circumstances.

One reviewer is still one reviewer, which is why every entry links to the narration behind it. A citation you can open is worth more than a credential you have to trust.

How errors get corrected

Every factual error found in the Islamic content goes on the corrections page with the date it was fixed, what was wrong as it had been published, and what it says now. Entries stay up. If a correction later needs correcting, that gets its own entry and the first one stays where it is.

That log opened on 5 August 2026 and it's empty. An empty log this early says the site is new. It doesn't say the site is free of errors.

Typos, broken links, layout problems and new features aren't corrections and don't go there. Padding the log with spelling fixes would make the site look more careful than it is.

Errors go to help@tasbih.info. The corrections page states what happens next: a reply within 7 days, and for a clear-cut fix, the change live within 14 days with the log entry going up the same day as the fix.

• A wrong collection, a wrong hadith number, or a link that lands on a different narration.
• A wrong grading, including dropping one that differs from the grade sunnah.com prints.
• A count shown as narrated when no narration gives that number.
• An error in the Arabic: a missing shadda, a wrong hamza, a dropped or doubled word.
• A mistranslation, or an English wording that shifts the meaning.
• A source we cited that doesn't say what we said it says.

The data is published, so you can check it without the app

Every phrase the counter uses is published as JSON under CC BY 4.0, and the dataset page explains the fields. It's generated from the same file the app itself reads, so the published data and the running product can't drift apart.

Each record carries narratedCount, which is null wherever no number is narrated, plus the count note, the grading, the collection and the source URL. The 5 editorial rules are in the file too, so anyone using the data gets the rules with it.

The post-salah program's declared gap is in the JSON as well. A developer building on this dataset inherits the missing Ayat al-Kursi step and the reason, rather than a list that looks finished.

Publishing it is itself the audit. Anyone can open the file, follow 17 hadith links to sunnah.com and 4 Qur'anic links to Quran.com, and check every reference, count and grading without installing the app or believing a word on this page.