feat(security): supply-chain advisory checker + lazy-install framework + tiered install fallback (#24220)

* feat(security): supply-chain advisory checker + lazy-install framework + tiered install fallback

Three coordinated mitigations for the Mini Shai-Hulud worm hitting
mistralai 2.4.6 on PyPI (2026-05-12) and for the next single-package
compromise that follows.

# What this PR makes true

1. Users with the poisoned mistralai 2.4.6 in their venv get a loud
   detection banner with copy-pasteable remediation steps the moment
   they run hermes (and on every gateway startup).
2. One quarantined / yanked PyPI package can no longer silently demote
   a fresh install to 'core only' — the installer keeps every other
   extra and tells the user which tier landed.
3. Future opt-in backends (Mistral, ElevenLabs, Honcho, etc.) can
   lazy-install on first use under a strict allowlist, instead of
   eagerly pulling everything at install time.

# Detection: hermes_cli/security_advisories.py

- ADVISORIES catalog (one entry currently: shai-hulud-2026-05 for
  mistralai==2.4.6). Adding the next one is a single dataclass.
- detect_compromised() uses importlib.metadata.version() — no pip
  dependency, works in uv venvs that lack pip.
- Banner cache (~/.hermes/cache/advisory_banner_seen) rate-limits
  the startup banner to once per 24h per advisory.
- Acks persisted to security.acked_advisories in config.yaml; never
  re-banner after ack.
- Wired into:
  * hermes doctor — runs first, prints full remediation block
  * hermes doctor --ack <id> — dismisses an advisory
  * cli.py interactive run() and single-query branches — short
    stderr banner pointing at hermes doctor
  * gateway/run.py startup — operator-visible warning in gateway.log

# Lazy-install framework: tools/lazy_deps.py

- LAZY_DEPS allowlist maps namespaced feature keys (tts.elevenlabs,
  memory.honcho, provider.bedrock, etc.) to pip specs.
- ensure(feature) installs missing deps in the active venv via the
  uv → pip → ensurepip ladder (matches tools_config._pip_install).
- Strict spec safety regex rejects URLs, file paths, shell metas,
  pip flag injection, control chars — only PyPI-by-name accepted.
- Gated on security.allow_lazy_installs (default true) plus the
  HERMES_DISABLE_LAZY_INSTALLS env var for restricted/audited envs.
- Migrated three backends as proof of pattern:
  * tools/tts_tool.py — _import_elevenlabs() calls ensure first
  * plugins/memory/honcho/client.py — get_honcho_client lazy-installs
  * tts.mistral / stt.mistral entries pre-registered for when PyPI
    restores mistralai

# Installer fallback tiers

scripts/install.sh, scripts/install.ps1, setup-hermes.sh:

- Centralised _BROKEN_EXTRAS list (currently: mistral). Edit one
  array when a transitive breaks; users keep every other extra.
- New 'all minus known-broken' tier between [all] and the existing
  PyPI-only-extras tier. Only kicks in when [all] fails resolve.
- All three tiers explicit: every fallback announces which tier
  landed and prints a re-run hint when not on Tier 1.
- install.ps1 and install.sh both regenerate their tier specs from
  the same _BROKEN_EXTRAS array so updates stay in sync.

Side effect: install.ps1 Tier 2 spec previously hardcoded 'mistral'
in its extra list — bug fixed by the refactor (mistral is filtered
out).

# Config

hermes_cli/config.py — DEFAULT_CONFIG.security gains:
- acked_advisories: []  (advisory IDs the user has dismissed)
- allow_lazy_installs: True  (security gate for ensure())

No config version bump needed — both keys nest under existing
security: block, and load_config's deep-merge picks up DEFAULT_CONFIG
defaults for users with older configs.

# Tests

tests/hermes_cli/test_security_advisories.py — 23 tests covering:
- detect_compromised matches/non-matches, wildcard frozenset
- ack persistence, idempotence, blank rejection, config-failure path
- banner cache rate limiting + 24h re-banner + ack-stops-banner
- short_banner_lines / full_remediation_text / render_doctor_section /
  gateway_log_message
- shipped catalog well-formedness invariant

tests/tools/test_lazy_deps.py — 40 tests covering:
- spec safety: 11 safe parametrized + 18 unsafe parametrized
- allowlist: unknown-feature rejection, namespace.name shape,
  every shipped spec passes the safety regex
- security gating: config flag, env var, default, fail-open
- ensure() happy/sad paths: already-satisfied, install success,
  pip stderr surfaced on failure, install-succeeds-but-still-missing
- is_available, feature_install_command

Combined: 63 new tests, all passing under scripts/run_tests.sh.

# Validation

- scripts/run_tests.sh tests/hermes_cli/test_security_advisories.py
  tests/tools/test_lazy_deps.py → 63/63 passing
- scripts/run_tests.sh tests/hermes_cli/test_doctor.py
  tests/hermes_cli/test_doctor_command_install.py
  tests/tools/test_tts_mistral.py tests/tools/test_transcription_tools.py
  tests/tools/test_transcription_dotenv_fallback.py → 165/165 passing
- scripts/run_tests.sh tests/hermes_cli/ tests/tools/ →
  9191 passed, 8 pre-existing failures (verified on origin/main
  before this change)
- bash -n on install.sh and setup-hermes.sh → OK
- py_compile on all modified .py files → OK
- End-to-end smoke test of detect_compromised + render_doctor_section
  + gateway_log_message with mocked installed version → produces
  copy-pasteable remediation output

# Community

Full advisory + remediation steps:
website/docs/community/security-advisories/shai-hulud-mistralai-2026-05.md

Short-form post drafts (Discord, GitHub pinned issue, README banner):
scripts/community-announcement-shai-hulud.md

Refs: PR #24205 (mistral disabled), Socket Security advisory
<https://socket.dev/blog/mini-shai-hulud-worm-pypi>

* build(deps): pin every direct dep to ==X.Y.Z (no ranges)

Companion to the supply-chain advisory work: replace every >=/</~= range
in pyproject.toml's [project.dependencies] and [project.optional-dependencies]
with an exact ==X.Y.Z pin sourced from uv.lock.

Why: ranges allow PyPI to ship a fresh version of any direct dep at any
time without a code review on our side. With ranges, the malicious
mistralai 2.4.6 release would have been pulled by every fresh
'pip install -e .[all]' for the hours between upload and PyPI's
quarantine — exactly the install window we got hit on. Exact pins close
that window: the only way a new package version reaches a user is via
an intentional update on our end.

What the user-facing change is: nothing, behavior-wise. Every package
resolves to the same version it was already resolving to via uv.lock —
the pins just remove the resolver's freedom to pick a different one.

Cost: any user installing Hermes alongside another package that requires
a newer pin gets a resolver conflict. Acceptable for our isolated-venv
install path; documented in the new comment block.

Build-system requires line (setuptools>=61.0) is intentionally left
as a range — pinning the build backend would block fresh pip from
bootstrapping the build on architectures where that exact wheel isn't
available.

mistral extra (mistralai==2.3.0) is pinned but stays out of [all]
(per PR #24205). 'uv lock' regeneration will fail until PyPI restores
mistralai; lockfile regeneration is gated behind that, NOT on every PR.

LAZY_DEPS in tools/lazy_deps.py also moved to exact pins so the lazy-
install pathway can never resolve a different version than the one
declared in pyproject.toml.

Validation:

- Cross-checked all 77 pinned direct deps in pyproject.toml against
  uv.lock — every pin matches the resolved version exactly.
- Cross-checked all LAZY_DEPS specs against uv.lock — same.
- 'uv pip install -e .[all] --dry-run' resolves 205 packages cleanly.
- tests/tools/test_lazy_deps.py + tests/hermes_cli/test_security_advisories.py
  → 63/63 passing (every shipped spec passes the safety regex).
- Doctor + TTS + transcription targeted suite → 146/146 passing.

* build(deps): hash-verify transitives via uv.lock; remove unresolvable [mistral] extra

You asked: 'what about the dependencies the dependencies rely on?' —
correctly noting that exact-pinning direct deps in pyproject.toml does
NOT cover the transitive graph. `pip install` and `uv pip install` both
re-resolve transitives fresh from PyPI at install time, so a compromised
transitive (e.g. `httpcore` if it got worm-poisoned tomorrow) would
still hit our users even with every direct dep exact-pinned.

# What this commit fixes

1. **Both real installer scripts now prefer `uv sync --locked` as Tier 0.**
   uv.lock records SHA256 hashes for every transitive — a compromised
   package with a different hash gets REJECTED. Falls through to the
   existing `uv pip install` cascade if the lockfile is missing or
   stale, with a loud warning that the fallback path does NOT
   hash-verify transitives. Previously only `setup-hermes.sh` (the dev
   path) used the lockfile; `scripts/install.sh` and `scripts/install.ps1`
   (the paths fresh users actually run) skipped it.

2. **Removed the `[mistral]` extra entirely.** The `mistralai` PyPI
   project is fully quarantined right now — every version returns 404,
   so any pin we wrote was unresolvable, which broke `uv lock --check`
   in CI. Restoration is documented in pyproject.toml as a 5-step
   checklist (verify, re-add extra, re-enable in 4 modules, regenerate
   lock, optionally re-add to [all]).

3. **Regenerated uv.lock.** 262 packages, mistralai/eval-type-backport/
   jsonpath-python pruned. `uv lock --check` now passes.

# Defense-in-depth view

| Layer                      | Where             | Protects against                          |
|----------------------------|-------------------|-------------------------------------------|
| Exact pins in pyproject    | direct deps       | new mistralai 2.4.6-style direct compromise |
| uv.lock + `--locked` install | transitive graph  | transitive worm injection                  |
| Tier-0 hash-verified path  | install.sh / .ps1 | actually USE the lockfile in fresh installs |
| `uv lock --check` CI gate  | every PR          | drift between pyproject and lockfile      |
| `hermes_cli/security_advisories.py` | runtime  | cleanup for users who already got hit      |

The exact pinning + hash verification together close the supply-chain
gap. Without the lockfile path, exact pins alone are theater.

# Validation

- `uv lock --check` → passes (262 packages resolved, no drift).
- `bash -n` on install.sh + setup-hermes.sh → OK.
- 209/209 tests passing across new + adjacent test files
  (test_lazy_deps.py, test_security_advisories.py, test_doctor.py,
  test_tts_mistral.py, test_transcription_tools.py).
- TOML parse OK.

* chore: remove community announcement drafts (PR body covers it)

* build(deps): lazy-install every opt-in backend (anthropic, search, terminal, platforms, dashboard)

Extends the lazy-install framework to cover everything that's not used by
every hermes session. Base install drops from ~60 packages to 45.

Moved out of core dependencies = []:
- anthropic   (only when provider=anthropic native, not via aggregators)
- exa-py, firecrawl-py, parallel-web (search backends; only when picked)
- fal-client  (image gen; only when picked)
- edge-tts    (default TTS but still optional)

New extras in pyproject.toml: [anthropic] [exa] [firecrawl] [parallel-web]
[fal] [edge-tts]. All added to [all].

New LAZY_DEPS entries: provider.anthropic, search.{exa,firecrawl,parallel},
tts.edge, image.fal, memory.hindsight, platform.{telegram,discord,matrix},
terminal.{modal,daytona,vercel}, tool.dashboard.

Each import site now calls ensure() before importing the SDK. Where the
module had a top-level try/except (telegram, discord, fastapi), the
graceful-fallback pattern was extended to lazy-install on first
check_*_requirements() call and re-bind module globals.

Updated test_windows_native_support.py tzdata check from snapshot
(>=2023.3 literal) to invariant (any version + win32 marker).

Validation:
- Base install: 45 packages (was ~60); 6 newly-extracted packages absent
- uv lock --check: passes (262 packages, no drift)
- 209/209 lazy_deps + advisory + doctor + tts/transcription tests passing
- py_compile clean on all 12 modified modules
This commit is contained in:
Teknium 2026-05-12 01:02:25 -07:00 committed by GitHub
parent 99ad2d1372
commit c1eb2dcda7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
28 changed files with 2433 additions and 243 deletions

View file

@ -0,0 +1,330 @@
"""Tests for hermes_cli.security_advisories.
The advisory module is the user-facing detection / remediation surface
for supply-chain attacks (e.g. the Mini Shai-Hulud worm of May 2026 that
poisoned mistralai 2.4.6 on PyPI). These tests exercise the public API in
isolation no real package metadata, no real config, no real cache.
"""
from __future__ import annotations
import time
from pathlib import Path
from typing import Iterator
import pytest
import hermes_cli.security_advisories as adv
# ---------------------------------------------------------------------------
# Fixtures
# ---------------------------------------------------------------------------
@pytest.fixture
def fake_advisory() -> adv.Advisory:
"""A self-contained Advisory used across tests."""
return adv.Advisory(
id="test-advisory-2026-99",
title="Test advisory",
summary="Pretend this package has been compromised.",
url="https://example.com/advisory",
compromised=(
("fake-malicious-pkg", frozenset({"6.6.6"})),
),
remediation=(
"pip uninstall -y fake-malicious-pkg",
"Rotate any credentials that may have been exposed.",
),
published="2026-01-01",
severity="critical",
)
@pytest.fixture
def isolated_home(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> Path:
"""Redirect HERMES_HOME so banner cache and config writes are sandboxed."""
home = tmp_path / ".hermes"
home.mkdir()
(home / "cache").mkdir()
monkeypatch.setattr(Path, "home", lambda: tmp_path)
monkeypatch.setenv("HERMES_HOME", str(home))
return home
@pytest.fixture
def patched_version(monkeypatch: pytest.MonkeyPatch) -> Iterator[dict[str, str]]:
"""Override _installed_version with a controllable lookup table."""
table: dict[str, str] = {}
monkeypatch.setattr(adv, "_installed_version", lambda pkg: table.get(pkg))
yield table
# ---------------------------------------------------------------------------
# detect_compromised
# ---------------------------------------------------------------------------
class TestDetectCompromised:
def test_no_match_returns_empty_list(self, fake_advisory, patched_version):
# No matching package installed.
hits = adv.detect_compromised(advisories=[fake_advisory])
assert hits == []
def test_exact_version_match(self, fake_advisory, patched_version):
patched_version["fake-malicious-pkg"] = "6.6.6"
hits = adv.detect_compromised(advisories=[fake_advisory])
assert len(hits) == 1
assert hits[0].advisory.id == fake_advisory.id
assert hits[0].package == "fake-malicious-pkg"
assert hits[0].installed_version == "6.6.6"
def test_safe_version_does_not_match(self, fake_advisory, patched_version):
# Package is installed but the version is not in the compromised set.
patched_version["fake-malicious-pkg"] = "6.6.5"
hits = adv.detect_compromised(advisories=[fake_advisory])
assert hits == []
def test_empty_compromised_set_matches_any_version(
self, patched_version
):
# An advisory with an empty version set is a "any version is suspect"
# wildcard — used when an entire maintainer namespace is owned.
wildcard = adv.Advisory(
id="wildcard",
title="Whole namespace owned",
summary="x",
url="x",
compromised=(("evil-namespace", frozenset()),),
remediation=("uninstall it",),
)
patched_version["evil-namespace"] = "0.0.1"
hits = adv.detect_compromised(advisories=[wildcard])
assert len(hits) == 1
assert hits[0].installed_version == "0.0.1"
# ---------------------------------------------------------------------------
# Acknowledgement persistence
# ---------------------------------------------------------------------------
class TestAck:
def test_get_acked_ids_empty_when_no_config(self, monkeypatch):
# load_config raises → returns empty set, doesn't crash.
monkeypatch.setattr(
"hermes_cli.config.load_config",
lambda: (_ for _ in ()).throw(RuntimeError("boom")),
)
assert adv.get_acked_ids() == set()
def test_filter_unacked_strips_dismissed(self, fake_advisory, monkeypatch):
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
monkeypatch.setattr(adv, "get_acked_ids", lambda: {fake_advisory.id})
assert adv.filter_unacked([hit]) == []
def test_filter_unacked_passes_through_unknown(
self, fake_advisory, monkeypatch
):
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
monkeypatch.setattr(adv, "get_acked_ids", lambda: set())
assert adv.filter_unacked([hit]) == [hit]
def test_ack_advisory_persists_id(self, isolated_home, monkeypatch):
# Stub the config layer end-to-end with a tiny in-memory store so we
# don't depend on the full hermes_cli.config bootstrap.
store: dict = {"security": {}}
monkeypatch.setattr(
"hermes_cli.config.load_config", lambda: store
)
monkeypatch.setattr(
"hermes_cli.config.save_config",
lambda cfg: store.update(cfg) or None,
)
assert adv.ack_advisory("test-advisory-2026-99") is True
assert "test-advisory-2026-99" in store["security"]["acked_advisories"]
# Idempotent.
adv.ack_advisory("test-advisory-2026-99")
assert (
store["security"]["acked_advisories"].count("test-advisory-2026-99")
== 1
)
def test_ack_advisory_rejects_blank(self, isolated_home):
assert adv.ack_advisory("") is False
assert adv.ack_advisory(" ") is False
# ---------------------------------------------------------------------------
# Banner cache rate limiting
# ---------------------------------------------------------------------------
class TestBannerCache:
def test_first_call_returns_due_hits(
self, fake_advisory, isolated_home, monkeypatch
):
monkeypatch.setattr(adv, "get_acked_ids", lambda: set())
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
due = adv.hits_due_for_banner([hit])
assert due == [hit]
def test_second_call_within_window_suppresses(
self, fake_advisory, isolated_home, monkeypatch
):
monkeypatch.setattr(adv, "get_acked_ids", lambda: set())
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
adv.hits_due_for_banner([hit])
# Same banner inside repeat window → suppressed.
again = adv.hits_due_for_banner([hit])
assert again == []
def test_call_after_window_re_banners(
self, fake_advisory, isolated_home, monkeypatch
):
monkeypatch.setattr(adv, "get_acked_ids", lambda: set())
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
adv.hits_due_for_banner([hit])
# Backdate the cache so it looks like the banner was shown more
# than 24h ago — should re-banner.
cache_path = adv._banner_cache_path()
assert cache_path is not None
old_lines = cache_path.read_text(encoding="utf-8").splitlines()
backdated = []
for line in old_lines:
parts = line.split(None, 1)
if len(parts) == 2:
backdated.append(f"{parts[0]} {time.time() - 48 * 3600}")
cache_path.write_text("\n".join(backdated) + "\n", encoding="utf-8")
again = adv.hits_due_for_banner([hit])
assert again == [hit]
def test_acked_hits_never_banner(
self, fake_advisory, isolated_home, monkeypatch
):
monkeypatch.setattr(adv, "get_acked_ids", lambda: {fake_advisory.id})
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
assert adv.hits_due_for_banner([hit]) == []
# ---------------------------------------------------------------------------
# Rendering
# ---------------------------------------------------------------------------
class TestRendering:
def test_short_banner_lines_includes_id_and_version(self, fake_advisory):
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
lines = adv.short_banner_lines([hit])
joined = "\n".join(lines)
assert fake_advisory.id in joined
assert fake_advisory.title in joined
assert "fake-malicious-pkg==6.6.6" in joined
assert "hermes doctor" in joined
def test_full_remediation_text_contains_all_steps(self, fake_advisory):
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
body = "\n".join(adv.full_remediation_text(hit))
# All remediation steps must be present.
for step in fake_advisory.remediation:
assert step in body
assert fake_advisory.url in body
assert fake_advisory.summary in body
def test_render_doctor_section_clean_state(self):
# No hits → success message, has_problems=False.
has_problems, lines = adv.render_doctor_section([])
assert has_problems is False
assert any("No active security advisories" in line for line in lines)
def test_render_doctor_section_with_unacked_hit(
self, fake_advisory, monkeypatch
):
monkeypatch.setattr(adv, "get_acked_ids", lambda: set())
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
has_problems, lines = adv.render_doctor_section([hit])
assert has_problems is True
body = "\n".join(lines)
assert fake_advisory.title in body
def test_gateway_log_message_singular(self, fake_advisory, monkeypatch):
monkeypatch.setattr(adv, "get_acked_ids", lambda: set())
hit = adv.AdvisoryHit(
advisory=fake_advisory,
package="fake-malicious-pkg",
installed_version="6.6.6",
)
msg = adv.gateway_log_message([hit])
assert msg is not None
assert fake_advisory.id in msg
assert "fake-malicious-pkg==6.6.6" in msg
def test_gateway_log_message_returns_none_for_no_hits(self):
assert adv.gateway_log_message([]) is None
# ---------------------------------------------------------------------------
# Real catalog smoke test
# ---------------------------------------------------------------------------
class TestRealCatalog:
def test_advisories_well_formed(self):
"""Every shipped advisory must be self-consistent.
Catches data-entry mistakes (empty IDs, missing remediation, bad
compromised tuples) before they ship.
"""
seen_ids: set[str] = set()
for advisory in adv.ADVISORIES:
assert advisory.id, "advisory has empty id"
assert advisory.id not in seen_ids, f"duplicate id {advisory.id}"
seen_ids.add(advisory.id)
assert advisory.title, f"{advisory.id}: empty title"
assert advisory.summary, f"{advisory.id}: empty summary"
assert advisory.remediation, f"{advisory.id}: empty remediation"
assert advisory.url.startswith("http"), \
f"{advisory.id}: bad url {advisory.url!r}"
assert advisory.compromised, \
f"{advisory.id}: empty compromised tuple"
for pkg, versions in advisory.compromised:
assert pkg, f"{advisory.id}: empty package name"
assert isinstance(versions, frozenset), \
f"{advisory.id}: versions must be frozenset"