CVE
Phoenix remote data leak via format string in applyChatTemplate
Affects arize-phoenix 13.0.0 through 14.8.0 when the applyChatTemplate GraphQL query endpoint is used. CVSS rates it critical and no vendor severity is given. Fixed in 15.7.0.
Written by anthropic/claude-haiku-5.5 from the sources below. Verify against them.
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:L
- Exploited
- No known exploitation
- Fix
- Fixed in 15.7.0
- Who is exposed
- Default setups
Do first: Check every install: default setups are affected.
- Last modified
- Whether it needs action now or is a reference record, by fixed rules, with why. Last checked is when cve.now last re-read it.What do these mean? →
- Reference
- Last checked
- A 0-100 score for how much a CVE matters across the industry: exploitation, exposure, how widely the software is used, severity and whether a fix exists.What do these mean? →
- 32/100 · provisional
- How many real-world setups can hit the bug, from the vendor's advisory and CISA data. It does not know your setup.What do these mean? →
- Default setups
- Whether working attack code is public (Exploit-DB, Metasploit, Nuclei or GitHub). Public code makes attacks easier.What do these mean? →
- None on Exploit-DB, Metasploit or GitHub
Summary
The Phoenix AI observability and evaluation platform has a format string injection flaw in its applyChatTemplate GraphQL query that lets crafted templates traverse Python internals. An attacker can exploit it remotely without logging in to read environment variables and internal module state, including PHOENIX_SECRET, which can enable administrative session-token forgery and decryption of protected credentials.
Original description
Phoenix is an AI observability and evaluation platform. From 13.0.0 until 15.7.0, the applyChatTemplate GraphQL query passed attacker-controlled F-string templates to FStringTemplateFormatter._format, while DotDict.__getattr__ did not prevent traversal through Python private attributes. A crafted field could traverse object internals such as __class__, __init__, and __globals__ to read process environment variables and internal module state. The endpoint is unauthenticated on default installations, and any account can reach it when authentication is enabled. Exposure of PHOENIX_SECRET can allow administrative session-token forgery and decryption of credentials protected by keys derived from that secret. This issue is fixed in version 15.7.0.
Who should check
Teams running arize-phoenix from 13.0.0 through 14.8.0, especially deployments exposing the applyChatTemplate GraphQL endpoint over a network.
“affecting all releases from v13.0.0 through v14.8.0 (current latest)”
cve.now Index
- Exploitation?/40Unknown. No CISA SSVC exploitation rating yet
- Exposure22/25Default setups, reachable over the network without a login
- Reach?/20Unknown. No tracked technology or package matched
- Severity10/10CVSS 10.0
- No patch0/5No signal
What would change the score
- Would score about 72 if added to CISA KEV
- Would score about 67 if CISA's SSVC rates exploitation active
What to do
Check every install: default setups are affected.
Upgrade to a fixed version: 15.7.0.
Check your systems with an agent
Paste this into a coding or ops agent. It checks your systems read-only and reports a table per asset.
Check whether my systems are affected by CVE-2026-53998 (Phoenix: Format String Injection in applyChatTemplate), using the cve.now data below.
Rules: work read-only. Do not run exploit code, proof-of-concept scripts or scanners that send attack payloads, and do not change any configuration, package, image or deployment. Report findings only.
What cve.now has for CVE-2026-53998 (entry last updated 2026-11-16):
Affected:
- Arize-ai phoenix: from 13.0.0, fixed in 15.7.0
Fixed versions: 15.7.0.
Conditions (all must be true for the bug to be hit):
- Affects arize-phoenix releases from 13.0.0 through 14.8.0 as stated in the advisory. (source: https://github.com/Arize-ai/phoenix/security/advisories/GHSA-gfv3-m769-5xm3)
- Requires use of the applyChatTemplate GraphQL query endpoint, which renders attacker-supplied templates. (source: https://github.com/Arize-ai/phoenix/security/advisories/GHSA-gfv3-m769-5xm3)
Setup: the bug is reachable in a default setup; no special feature or setting is needed.
Attacker needs: network access, no login, no user action.
CISA KEV: CVE-2026-53998 is not listed as of cve.now's last read.
This KEV status is for CVE-2026-53998 only. If you also check another CVE, look up its own KEV status; never assume it from this one.
How to check:
- Installed software: check what is actually running, not what a repo declares. Read the version from each host, container image and Kubernetes node (the binary's version output, the OS package database, image layers), and check whether the conditions above hold there. Distribution packages may carry a backported fix under an older version number; confirm against the distribution's own advisory.
Verify against the linked advisories before concluding, and cite the date you read each one:
- https://github.com/Arize-ai/phoenix/security/advisories/GHSA-gfv3-m769-5xm3
- https://github.com/Arize-ai/phoenix/pull/13197
- https://github.com/Arize-ai/phoenix/commit/e7224a685589f2ffae1fefcc68ff210bd6ad28b8
- https://github.com/Arize-ai/phoenix/releases/tag/arize-phoenix-v15.7.0
Do not conclude an asset is unaffected unless you checked it and the evidence covers both its version and the conditions above. If you could not check something, mark it unknown and say what is missing.
Report a table with one row per asset and these columns: Asset | Version or build | Status (affected, fixed, not applicable, unknown) | Compensating controls | Exposure (internet-facing, internal, none) | Evidence | Next action.
Entry: https://cve.now/CVE-2026-53998
For follow-up questions, use the cve.now MCP server at https://mcp.cve.now/mcp (no authentication), for example execute with name cve.get and arguments {"id": "CVE-2026-53998"}.Timeline
- CVE published
- First seen by cve.now
- Posted on X by cve.now
Affected versions
- Arize-ai phoenix
From 13.0.0 · fixed in 15.7.0
Evidence 6
Vendor advisory · 1
Establishes: Affected and fixed versions.
- Advisory
CVE record · 2
Establishes: The identifier and description.
- Record
- Record
Other links · 3
- Repository
- Repository
- Repository
Corrections
None so far.
Something wrong on this page? Report an error. Every fix is listed here.
Questions about CVE-2026-53998
Is CVE-2026-53998 being exploited?
It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.Source: cve.now feed data
What does this mean?
The applyChatTemplate GraphQL endpoint renders attacker-supplied templates without safe handling of format strings. This lets a request read server environment variables, internal module state, and secrets such as PHOENIX_SECRET, the key used to sign JWTs.Source: github.com
Which versions are affected by CVE-2026-53998?
phoenix: from 13.0.0 before 15.7.0.Source: cve.now feed data
Am I affected, and how do I check?
The advisory states that releases v13.0.0 through v14.8.0 are affected. The setup condition is use of the applyChatTemplate GraphQL query endpoint, so check the installed arize-phoenix version and whether that endpoint is in use.Source: github.com
How do I fix CVE-2026-53998?
Upgrade arize-phoenix to 15.7.0 or later. The facts do not list vendor patch identifiers for this issue.Source: github.com