Provenance Record

Well-structured News piece (Overview / What We Know / What We Don't Know / Analysis format). Neutral, attribution-heavy prose throughout — nearly every factual claim carries an inline 'according to X' or 'the statement said' hedge, which is appropriate for a story built entirely on one party's allegations against a named individual. 678 words, within the 400-1200 News range.

Personally read all 5 cited sources. 4 of 5 have local gzip snapshots in sources/2026-07/openmandriva-says-ex-contributor-deleted-years-of-repository-work-and-sabotaged-its-cooker-branch/ whose sha256 I independently recomputed and confirmed match manifest.json: source-0.html.gz (forum.openmandriva.org statement, matches 6a5b889d...), source-2.html.gz (theregister.com, matches 3cfcfbed...), source-3.html.gz (linuxiac.com, matches 801d782b...), source-4.html.gz (fossforce.com, matches c0dc8e68...). The BleepingComputer source (index 1) returned HTTP 403 (bot-blocked) with file:null in the manifest, so per the mandatory fallback procedure I WebFetched the live URL directly and confirmed its content. Every direct quote in the article was checked verbatim against its source snapshot/fetch: the AngryPenguin forum quotes ('our distribution has experienced several disruptions...', 'The team had no hesitation in trusting him...', 'began to behave in abusive ways...', 'I decided to take action and kicked the attacker out', 'including Davide, a friend of the attacker', 'This infuriated Davide so much that...', 'deleted part of our repository from GitHub...', 'also decided to publish an empty package...obsoleted all gnome and cosmic packages', 'could have damaged the systems of people using gnome or cosmic') all appear verbatim in source-0. The Register quote ('repeated abusive behavior toward users and project members, much of it in private messages') and the 'appears to have been confined to bleeding-edge users' characterization both appear verbatim in source-2. The 'We understand that Davide's actions were unacceptable...' quote appears verbatim in both source-0 (forum) and source-4 (FOSS Force); the article cites FOSS Force, which is accurate since FOSS Force does reproduce it verbatim. Beatrici's rebuttal quotes ('Let me state right away that this was by no means a sabotage', 'The objective was not to harm the distribution I cared for and contributed to for the past 3 years') were confirmed verbatim via live WebFetch of BleepingComputer.

No fabricated facts or hallucinated quotes found. Two attribution-precision issues identified: (1) the article quotes 'no other violations' and cites Linuxiac, but Linuxiac's own text paraphrases this as 'found no other problems aside from the removed packages' — the verbatim phrase 'no other violations' only appears in the original forum statement (source-0), not in Linuxiac's article. The underlying fact is true and multiply-sourced, but the citation for that exact wording points to the wrong outlet. (2) The article presents 'triggered by a few members of the project who did not agree with OpenMandriva's focus on KDE and LXQt' as a direct quote attributed to Beatrici ('He said the underlying conflict was...'), but per BleepingComputer's own text this sentence is the reporter's indirect/reported-speech paraphrase ('As Beatrici says, this action was triggered by...'), not something Beatrici said in his own words inside quotation marks in the source. The wording is reproduced verbatim from BleepingComputer, so this is a quote-provenance/formatting issue rather than a fabrication — the substance (Beatrici's stated rationale) is accurately sourced to BleepingComputer.

Substantively strong, well-sourced, and unusually careful for a story about a personal/reputational dispute — every serious allegation is explicitly attributed to OpenMandriva's own statement or to named outlets, and the accused party's rebuttal is quoted at comparable length. Two minor, honestly-correctable attribution-precision issues (a quote wrongly cited to Linuxiac instead of the forum, and a reporter's paraphrase presented as a direct quote) do not affect the headline, summary, or lead and do not change the substance of any claim. APPROVE_WITH_CORRECTIONS is warranted; a REJECT would be disproportionate given both issues are single-sentence attribution nuances fully recoverable via a corrections note, and the article's central claims are all soundly sourced.