Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
41 changes: 38 additions & 3 deletions execution_trace.py
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,7 @@

import fcntl
import json
import re
import uuid
from contextlib import contextmanager
from datetime import datetime, timezone
Expand All @@ -18,6 +19,14 @@
TERMINAL_STATUSES = {"completed", "failed", "blocked"}


_TRACE_REDACTIONS = (
(re.compile(r"-----BEGIN [^-\r\n]*PRIVATE KEY-----[\s\S]*?-----END [^-\r\n]*PRIVATE KEY-----"), "[redacted]"),
(re.compile(r"(?i)\bBearer\s+[A-Za-z0-9._~+/=-]+"), "Bearer [redacted]"),
(re.compile(r"(?i)(?<![A-Za-z0-9])(?:gh[pousr]_[A-Za-z0-9_]+|github_pat_[A-Za-z0-9_]+|moltbook_[A-Za-z0-9_-]+)"), "[redacted]"),
(re.compile(r"(?<![/:A-Za-z0-9_])/(?:[A-Za-z0-9._~+@%-]+/)*[A-Za-z0-9._~+@%-]+"), "[local]"),
)


def now_iso() -> str:
return datetime.now(UTC).isoformat()

Expand All @@ -26,14 +35,31 @@ def default_trace() -> dict[str, Any]:
return {"schema_version": 1, "current": None, "history": []}


def _redact_trace_value(value: Any) -> Any:
if isinstance(value, str):
for pattern, replacement in _TRACE_REDACTIONS:
value = pattern.sub(replacement, value)
return value
if isinstance(value, dict):
return {
_redact_trace_value(key) if isinstance(key, str) else key: _redact_trace_value(item)
for key, item in value.items()
}
if isinstance(value, list):
return [_redact_trace_value(item) for item in value]
if isinstance(value, tuple):
return tuple(_redact_trace_value(item) for item in value)
return value


class ExecutionTrace:
"""A small cross-process event log shared by the live worker and Console."""

def __init__(self, state_path: Path, run_id: str | None = None, trace_path: Path | None = None):
self.state_path = Path(state_path).expanduser()
self.path = (trace_path or self.state_path / TRACE_FILE_NAME).expanduser()
self.lock_path = self.path.with_suffix(self.path.suffix + ".lock")
self.run_id = run_id
self.run_id = _redact_trace_value(run_id) if run_id else None

def _read_unlocked(self) -> dict[str, Any]:
if not self.path.exists():
Expand All @@ -48,7 +74,7 @@ def _read_unlocked(self) -> dict[str, Any]:
merged.update(value)
if not isinstance(merged.get("history"), list):
merged["history"] = []
return merged
return _redact_trace_value(merged)

@contextmanager
def _locked(self) -> Iterator[None]:
Expand All @@ -62,7 +88,8 @@ def _locked(self) -> Iterator[None]:

def _write_unlocked(self, value: dict[str, Any]) -> None:
temporary = self.path.with_suffix(self.path.suffix + ".tmp")
temporary.write_text(json.dumps(value, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
safe_value = _redact_trace_value(value)
temporary.write_text(json.dumps(safe_value, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
temporary.replace(self.path)

def snapshot(self) -> dict[str, Any]:
Expand All @@ -78,6 +105,11 @@ def _event(
command: str | None,
details: dict[str, Any] | None,
) -> dict[str, Any]:
phase = _redact_trace_value(phase)
status = _redact_trace_value(status)
message = _redact_trace_value(message)
command = _redact_trace_value(command) if command else None
details = _redact_trace_value(details) if details else None
events = list(current.get("events") or [])
event = {
"sequence": len(events) + 1,
Expand Down Expand Up @@ -108,6 +140,9 @@ def start(
trigger: str = "systemd timer",
message: str = "Maintainer cycle started",
) -> str:
kind = _redact_trace_value(kind)
trigger = _redact_trace_value(trigger)
message = _redact_trace_value(message)
self.run_id = self.run_id or f"{kind}-{uuid.uuid4().hex[:12]}"
started_at = now_iso()
current: dict[str, Any] = {
Expand Down
34 changes: 34 additions & 0 deletions tests/test_execution_trace.py
Original file line number Diff line number Diff line change
Expand Up @@ -44,3 +44,37 @@ def test_execution_trace_ignores_updates_from_a_previous_run(tmp_path):
assert snapshot["current"]["id"] == "second"
assert snapshot["current"]["events"][-1]["message"] != "stale update"
assert first_id != "second"


def test_execution_trace_redacts_sensitive_event_content_before_persistence(tmp_path):
bearer = "Bearer " + "bearer-example-value"
github_token = "gh" + "p_" + ("0123456789" * 3)
moltbook_token = "molt" + "book_sk_" + ("0123456789abcdef" * 2)
private_key_material = "private" + "-key-material"
private_key = "-----BEGIN " + "PRIVATE KEY-----\n" + private_key_material + "\n-----END " + "PRIVATE KEY-----"
local_path = str(tmp_path / "private" / "config.toml")

trace = ExecutionTrace(tmp_path / "state")
trace.start(message=f"started with {bearer}")
trace.event(
"authentication",
message=f"{bearer} {github_token} {moltbook_token}",
command=f"cat {local_path} --token {github_token}",
details={
"bearer": bearer,
"github": github_token,
"moltbook": moltbook_token,
"private_key": private_key,
"path": local_path,
"nested": [private_key, local_path],
},
)

persisted = (tmp_path / "state" / "execution.json").read_text(encoding="utf-8")
for secret in (bearer, github_token, moltbook_token, private_key_material, str(tmp_path)):
assert secret not in persisted
assert "[redacted]" in persisted
assert "[local]" in persisted

snapshot = trace.snapshot()
assert [event["phase"] for event in snapshot["current"]["events"]] == ["starting", "authentication"]