CVE-2025-59681:django: SQL injection via column alias
A historical CVE, retraced through the questions Lachesis asks of vulnerable code.
An attacker-controlled column alias reached the SQL query because Django's safety check missed one dangerous character.
That is the shape of the failure. The rest of this case file follows the evidence behind it.
What was known.
What Lachesis established.
Known before Lachesis ran
The CVE and vulnerable release were selected from public history. We are not claiming novel discovery.
- CVE
- CVE-2025-59681
- Vulnerable target
- 5.2.6 (git 75c4403) · vulnerable
- Prior knowledge
- CVE-2025-59681 · 5.2.6 (git 75c4403) is vulnerable
Independent re-detection
No source/sink/guard fed from memory. Hypothesis 'Q(_connector=)' from memory was REFUTED by the 5.2.6->5.2.7 diff, which showed the real fix is check_alias's FORBIDDEN_ALIAS_PATTERN gaining '#'. Drove the hunt on the diff + graph evidence only.
- Seeded inputs
- None
- Run timestamp
- 2026-08-24T12:06:09Z
- Evidence artifact
- ~/.lachesis/graphs/django_cve_2025_59681.kuzu
The commands and outputs below come from this recorded Lachesis run.
What Lachesis reconstructed.
One historical repository. Four captured queries. A complete source-to-sink argument.
Lachesis follows the same evidence path through unfamiliar repositories.
Here is how the path becomes visible.
These are the recorded questions Lachesis asked of the historical vulnerable code, followed by the raw result and source location each query returned.
Captured runThis block reflects the real blind drive of the graph we built for Django. We loaded the whole graph, listed every bug family with nothing chosen ahead of time, and the SQL injection sinks came out of that. The candidate, model tag and guard reading below are the facts captured on that drive on that drive. No query was seeded; the finding was rediscovered from graph structure, so the mode is independent-redetection.
Load the graph and list every bug family
First we load the graph and ask the tool to list every bug family, with nothing picked ahead of time. It reports all 8 domains and 31 sink constructors and says the census is complete for what the graph can observe. Injection is one row among all of them.
{"move": "load_graph", "graph": "/Users/riyandhiman/.lachesis/graphs/django_cve_2025_59681.kuzu", "profile": "all"}
CANDIDATE_CENSUS
move: candidate_census
taxonomy (8):
domain=lifecycle title=Resource lifecycle enumerable=True
domain=memory title=Memory safety enumerable=True
domain=injection title=Injection enumerable=True
domain=navigation title=Request forgery & redirection enumerable=True
domain=object-integrity title=Object integrity enumerable=True
domain=filesystem title=Filesystem enumerable=True
domain=crypto-config title=Cryptography & transport config enumerable=True
domain=resource title=Resource exhaustion enumerable=True
constructors (31): complete_for_observable_graph=True
applied: TrueA small line with a large consequence.
The character class omits '#', which starts an inline comment on MySQL and MariaDB, so a '#' alias slips past the check and reaches the SELECT clause.
FORBIDDEN_ALIAS_PATTERN = _lazy_re_compile(r"['`\"\]\[;\s]|--|/\*|\*/")
// Add '#' to the pattern: r"['`\"\]\[;\s]|#|--|/\*|\*/" (Django 5.2.7).
The original record.
Read the historical advisory and vulnerability record behind this reconstruction.