Sahel developments — 22 November 2026
Mali's general staff said on 18 November that an air and heliborne operation with unnamed partners freed 24 soldiers held near In-Tebezas, south-east of Kidal, and took them to Gao. Two days later RFI carried the FLA's account: the movement says it handed the men over at an agreed place, and RFI identifies the partners as Africa Corps from the corps' own video. The two accounts agree on the recovery, not on whether there was a fight. In Niger, RFI found that the Algerian escort of the 31 October fuel convoy went home with it and saw no movement at Madama: the deployment is still a decision and one delivery. Reuters reported an analysis identifying Africa Corps' use of Lancet loitering munitions in Mali since April. No attack on a position or a supply convoy was found.
The developments
Mali's army says it freed 24 soldiers near Kidal by assault; the FLA says it handed them over
Mali's general staff announced that 24 soldiers held by armed groups at In-Tebezas, about 110 km south-east of Kidal, were freed on Wednesday 18 November 2026 at about 13:00. By its communiqué, as relayed by Bénin Web TV and APA, air reconnaissance located the site, strikes hit the jailers' guard posts and troops landed by helicopter went in, with the support of unnamed partners. The 24 were taken to Gao, the force had no losses and fighters were said to be neutralised, with no figure. The communiqué named neither the captors nor how long the men had been held. On 20 November RFI published a different account. FLA commanders told it the movement had moved the soldiers to a place agreed in advance for the army to collect them, on orders from above, and that no FLA fighter was there when the Russians arrived. RFI says Africa Corps published video of the operation: a helicopter with Russian soldiers landing unopposed and collecting the men, with gunfire and explosions but no destroyed target or dead fighter shown. Sources on all sides told RFI there was no exchange. RFI also reports Malian and FLA representatives in Algiers for several days, with Algeria trying to restart talks; an FLA cadre said there was nothing to negotiate.
Why it matters. The accounts agree that 24 soldiers left captivity near In-Tebezas on 18 November and reached Gao, and that the army acted there with outside help, Africa Corps by RFI's reading of the corps' own video. They disagree on whether there was a fight, and that is the whole difference. On the army's account it is the first ground action reported in the Kidal countryside since the town was re-entered on 6 October. On the FLA's it is another transfer of captured soldiers, after the releases of August, at least one of which was an exchange, and a sign of contact between Bamako and the movement rather than of reach. Neither changes who holds what: no position was taken or put back, and Kidal's supply and administration are still not reported. An FLA communiqué, the Africa Corps video itself or an outcome of the Algiers contacts would separate the readings.
RFI published on 19 November a reconstruction of the first Algerian fuel convoy into Niger. The 53 tankers of the state company Naftal, carrying more than 1.6 million litres from the Sbaa refinery in Adrar, were photographed at the Arlit customs post on Friday 30 October and filmed about 24 hours later in Agadez, less than 5 km from Air Base 201. Algerian and Nigerien sources told RFI that the escort left with the convoy once it had unloaded. RFI's satellite imagery showed no movement at Madama, the former French base where press reports had placed the Algerian troops. A defence source said there was no unit there for now and that there could be an outpost; a source put a main base probably near Agadez. RFI also found false several items circulating about the mission: AI-generated images first posted on 5 November, a claimed strike on the convoy on 17–18 November spread on X, and a 2013 photograph from In Amenas.
Why it matters. It answers the gap the Watch question left open: whether any Algerian personnel stayed in Niger after the delivery. By these sources they did not, and the one Algerian military activity reported inside Niger remains the escort of 30–31 October. The authorisation voted on 25 October has produced no observed stationed unit, no published term and no second convoy. That is consistent with a mission still being prepared and with one that never grows; the defence source's Agadez base is an expectation, not an observation, and RFI's imagery covers one site, not the country. A named unit at a site, a second escorted convoy or a published term would move the record.
Reuters reported on 19 November an analysis by Conflict Insights Group, a US-based research organisation, which identified at least ten Africa Corps strikes in Mali since April 2026 in which a ZALA Lancet one-way attack drone was paired with a ZALA Aero surveillance drone. The group also reported that air and drone strikes in Mali, and the deaths recorded from them, rose after April; Vigil does not repeat those figures, which rest on a third-party event dataset the group processed. On at least one occasion, it found, a Lancet almost certainly hit a vehicle carrying civilians. Other conflict monitors, Reuters adds, report Shahed-type drones and their Russian versions used in Mali. No strike locations or dates were published, and the Russian and Malian governments did not respond.
Why it matters. It is the first published identification in Vigil's record of a named Russian loitering munition in Africa Corps' hands in Mali, the drone-led way of fighting that the corps' record already describes, made specific. In Ukraine the Lancet is the strike half of Russia's reconnaissance-strike pairing, as Vigil's equipment record describes it, so its use in Mali means the corps brings that method, not only gunships and road columns. What it establishes is use. It does not establish when or how the munitions arrived, how many there are, or where any strike fell, so it shows nothing about Russian deliveries to Mali or about dependence on them. Strike locations, or a delivery with a date, would.
Where this fits
This is one conflict's dated review for the week. The cross-conflict edition for the same week carries the most consequential findings from every conflict Vigil covers. The standing picture this review updates is the Sahel coverage. How Vigil grades, dates and sources its claims is set out in the methodology, and anything later found wrong is recorded in the corrections log.
By email Get Vigil’s Weekly Briefs and new Analytical Notes by email. What Vigil Dispatch sends
← Previous Sahel brief — 2026-11-15Every Sahel review All weekly briefs