CVE-2026-56834: PraisonAI dynamic-context artifact tools read arbitrary host files outside artifact storage
PraisonAI dynamic-context artifact tools read arbitrary host files outside artifact storage Summary PraisonAI's Dynamic Context Discovery feature exposes artifact helper tools through ctx.gettools(): The official documentation describes these helpers as a way for the agent to explore large tool-output artifacts that were queued by the middleware: - large tool outputs are saved as artifacts;
Published Jun 18, 2026 · Updated Jul 20, 2026
Affected packages
praisonai
- Ecosystem
- PyPI
- Affected
- >= 3.8.1, <= 4.6.58
- Fixed in
- 4.6.59
- Check
pip show praisonai
Advisory details
From the GitHub Advisory Database (CC BY 4.0).
PraisonAI dynamic-context artifact tools read arbitrary host files outside artifact storage
Summary
PraisonAI's Dynamic Context Discovery feature exposes artifact helper tools through ctx.get_tools():
ctx = setup_dynamic_context()
agent = Agent(
instructions="You are a data analyst.",
tools=ctx.get_tools(),
hooks=[ctx.get_middleware()],
)The official documentation describes these helpers as a way for the agent to explore large tool-output artifacts that were queued by the middleware:
- large tool outputs are saved as artifacts;
- the agent receives compact artifact references; and
- the agent uses
artifact_tailandartifact_grepto explore that data.
The implemented artifact tools do not enforce that the supplied artifact_path is an artifact created by the configured store or that it lives under the configured artifact base directory. Instead, artifact_head, artifact_tail, artifact_grep, and artifact_chunk wrap the caller-supplied path directly into an ArtifactRef and then read it from the host filesystem.
As a result, any prompt/user/tool-caller that can influence those tool arguments can read files readable by the PraisonAI process, such as project .env files, cloud credentials, SSH keys, source files, or other local data.
Affected Product
- Repository:
MervinPraison/PraisonAI - Ecosystem:
pip - Package:
praisonai - Component: Dynamic Context Discovery artifact tools
- Current source path:
src/praisonai/praisonai/context/queue.py - Artifact store path:
src/praisonai/praisonai/context/artifact_store.py - Latest PyPI version validated:
4.6.58 - Current
origin/mainvalidated:1ad58ca02975ff1398efeda694ea2ab78f20cf3e - Current
origin/maintag validated:v4.6.58
Suggested affected range:
pip:praisonai >= 3.8.1, <= 4.6.58Representative local sweep:
3.8.1: vulnerable4.0.0: vulnerable4.5.113: vulnerable4.6.33: vulnerable4.6.34: vulnerable4.6.40: vulnerable4.6.50: vulnerable4.6.58: vulnerable
Root Cause
create_artifact_tools() creates an artifact store bound to base_dir, but the read tools do not use base_dir for containment.
For example, artifact_head() accepts artifact_path and immediately creates an ArtifactRef with that path:
def artifact_head(artifact_path: str, lines: int = 50) -> str:
ref = ArtifactRef(path=artifact_path, summary="", size_bytes=0)
try:
return artifact_store.head(ref, lines=lines)
except FileNotFoundError:
return f"Error: Artifact not found: {artifact_path}"artifact_tail(), artifact_grep(), and artifact_chunk() have the same pattern. They trust the caller-supplied path rather than resolving it through an artifact identifier, store lookup, manifest, or base-directory containment check.
The store methods then read that path directly:
def head(self, ref: ArtifactRef, lines: int = 50) -> str:
file_path = Path(ref.path)
if not file_path.exists():
raise FileNotFoundError(f"Artifact not found: {ref.path}")
result_lines = []
with open(file_path, "r", encoding="utf-8", errors="replace") as f:
...There is no check equivalent to:
resolved = Path(ref.path).resolve()
base = self.base_dir.resolve()
resolved.relative_to(base)There is also no check that the file has a valid .meta sidecar or appears in artifact_list().
Local PoV
Run against the latest PyPI package:
uv run --with 'praisonai==4.6.58' \
python poc/pov_prai_cand_026_artifact_tools_arbitrary_file_read.py --jsonThe PoV:
- Creates a temporary artifact base directory.
- Creates a separate
outside-secret.txtfile outside that base directory. - Stores one legitimate artifact through
FileSystemArtifactStore.store(). - Calls
artifact_head()on the legitimate artifact as a positive control. - Calls
artifact_head(),artifact_grep(), andartifact_chunk()on the outside file path. - Confirms
artifact_list()does not list the outside file.
Observed output summary from evidence/pov-pypi-4.6.58.json:
{
"package": "praisonai",
"package_version": "4.6.58",
"controls": {
"outside_file_not_listed": true,
"outside_file_outside_base_dir": true,
"valid_artifact_read_works": true
},
"outside_head": "PRAI-CAND-026-OUTSIDE-ARTIFACT-SECRET",
"outside_grep": "Found 1 matches:\\n\\n--- Line 1 ---\\n> PRAI-CAND-026-OUTSIDE-ARTIFACT-SECRET\\n second line",
"outside_chunk": "PRAI-CAND-026-OUTSIDE-ARTIFACT-SECRET",
"outside_file_listed_by_artifact_list": false,
"vulnerable": true
}The PoV was rerun successfully after a fresh origin/main fetch; see evidence/pov-pypi-4.6.58-rerun.json.
The PoV is local-only. It does not start a server, contact a third-party target, or use real credentials.
Why This Is Not Intended Behavior
This report does not claim that every file-reading tool is automatically a vulnerability. The issue is narrower: tools documented and named as artifact helpers accept arbitrary host file paths.
The controls show the intended boundary:
- a valid artifact stored under
base_diris readable; - an outside file is not returned by
artifact_list(); - the outside file is outside
base_dir; and - the read helpers still disclose the outside file when handed its absolute path.
PraisonAI's own context-security documentation recommends relative paths and reviewing ignore rules to avoid sensitive-file exposure. Those controls are bypassed when artifact tools can be pointed directly at any readable host path.
Impact
If a PraisonAI application exposes an agent with ctx.get_tools() to untrusted or lower-trust prompts, the lower-trust caller can request artifact tools against arbitrary local paths. This can disclose sensitive host files readable by the PraisonAI process, including:
- project
.envfiles; - cloud or service credentials;
- SSH keys;
- local application configuration;
- source files and private data; and
- terminal/history artifacts from other runs if the path is known or guessed.
The impact is confidentiality-only in the tested surface. Integrity and availability are not claimed for this report.
Duplicate Posture
I checked visible PraisonAI advisories and local prior PraisonAI submissions. This is distinct from nearby file-read/file-write issues:
GHSA-9cr9-25q5-8prj/CVE-2026-47394covers MCP CLIworkflow.show,workflow.validate, anddeploy.validatepath handling. This report covers Dynamic Context Discovery artifact tools incontext/queue.py.GHSA-hvhp-v2gc-268q/CVE-2026-47397coverswrite_filearbitrary file write whenworkspace=None. This report is a read-only disclosure issue in artifact helper tools.- Public recipe registry path traversal advisories cover recipe publish/pull storage and extraction. This report does not involve the recipe registry.
- Local prior submissions in this harness do not cover
artifact_head,artifact_tail,artifact_grep,artifact_chunk, orFileSystemArtifactStorepath containment.
Severity
Suggested severity: High.
Suggested CVSS v3.1:
Rationale:
AV: applies when an application exposes a PraisonAI agent over a network chat/API surface, which is a documented PraisonAI deployment pattern.AC: no race, special environment, or complex path manipulation is required; an absolute readable path is sufficient.PR: an unauthenticated or public-facing agent endpoint can be exploited without an account. Deployments that require authenticated chat/API access may score this asPR:L.UI: the attacker directly supplies the prompt/tool argument to the exposed agent surface.C: arbitrary readable host files can contain secrets or private data.I/A: this report demonstrates read-only disclosure.
Remediation
Do not let artifact tools open arbitrary paths. Prefer stable artifact IDs over raw filesystem paths in tool arguments.
Recommended fixes:
- Change tool schemas to accept
artifact_idplus optionalrun_idandagent_id, then resolve those through the artifact store's metadata/index. - If path arguments must remain for compatibility, resolve the path with
Path(path).resolve()and reject it unless it is underartifact_store.base_dir.resolve(). - Require a valid artifact metadata sidecar for read helpers. Files not created by
FileSystemArtifactStore.store()should not be readable through artifact tools. - Apply the same containment check to
load(),head(),tail(),grep(),chunk(), anddelete(). - Avoid returning absolute host paths in prompt-visible artifact references when an opaque artifact ID would suffice.
Minimal containment helper:
def _resolve_artifact_path(self, path: str) -> Path:
resolved = Path(path).expanduser().resolve()
base = self.base_dir.resolve()
try:
resolved.relative_to(base)
except ValueError as exc:
raise PermissionError("Artifact path is outside artifact storage") from exc
return resolvedThis helper should be paired with metadata-sidecar validation so arbitrary non-artifact files placed under the base directory are not automatically treated as valid artifacts.
Credited to rexpository.
References
How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology
More PyPI vulnerabilities
All PyPI entriesCVE-2026-61438
PraisonAI vulnerable to Remote Code Execution via Broken AST Sandbox
high7.3Oct 29, 2026CVE-2026-60085
PraisonAI: SecurityPolicy command/path/import restrictions are completely unenforced by the default SubprocessSandbox backend
high8.7Oct 8, 2026CVE-2026-60086
PraisonAI: Prompt-injection defense blocks only when 3+ detector families fire simultaneously; realistic single-vector injections pass through unblocked
medium5.3Oct 8, 2026CVE-2026-60088
PraisonAI: Project custom command templates can read outside-workspace files into model prompts
medium5.5Oct 8, 2026CVE-2026-60090
PraisonAI: PGVector and Cassandra knowledge stores interpolate vector dimensions into DDL
mediumOct 8, 2026CVE-2026-60091
PraisonAI: Jobs API is unauthenticated by default and allows attacker-controlled webhook SSRF
high7.2Oct 8, 2026
