Raven-1/tests/test_package_skeleton.py
Handsome-wzw 7b45527b5b fix(tools): an array sent as its own JSON is still an array
## Summary

Reported as "deliver_files is broken". It is not: **the tool never ran.**

Every `deliver_files` call on one install was refused before dispatch -- five in a row in a single session -- and nothing was ever delivered:

```
Error: Invalid parameters for tool 'deliver_files': files should be array
```

What the model sent was the array, spelled as a string:

```json
{"files": "[{\"path\": \"...poster.png\", \"title\": \"...\"}]", "message": "..."}
```

`cast_params` already forgives that spelling for every scalar type -- `"3"` for an integer, `"true"` for a boolean -- and did not for arrays or objects, which are exactly the types a model is most likely to stringify.

## Type

- [x] Fix
- [ ] Feature

## The edge of the forgiving is the design

When the value is a string, try `json.loads`, and take the result **only if it is the type the schema asked for**. A string that is not JSON, or is JSON of the wrong shape, is returned unchanged so the validator still names the real problem -- the point is to widen what is accepted, not to swallow a type error.

**Nothing here can raise.** Turning "your argument has the wrong type" into a traceback would be worse than the refusal it replaces, so the parse is wrapped and failure falls through.

**A `type: "string"` field is never parsed.** JSON is ordinary text in a message body or a file's content, and replacing it would hand the tool something it never asked for. Two independent things prevent it today (the string branch returns before the parse is reached, and the parse only wins on a type match), which is why the case pinning it is a tripwire rather than a load-bearing assertion -- said so in its own docstring rather than left to look like coverage.

## Not just deliver_files

Six built-in tools declare an array or object parameter and were all exposed:

| tool | parameter |
|---|---|
| `deliver_files` | `files` |
| `ask_user` | `questions` |
| `image_generate` | `images` |
| `message` | `media` |
| `create_playbook` | `skills` |
| `video_generate` | `params` |

The same failure was already seen from the other side: `acp_client/acp_agent.py` records `update_task_state` and `operations should be array` from a real dispatch.

## Verification

```
make lint-python    ruff check + ruff format --check, 1850 files
make lint-imports   9 contracts kept, 0 broken
pytest tests/       19,699 passed, 7 failed (all pre-existing, see below)
```

- [x] Relevant tests pass locally
- [x] Relevant lint / type checks pass locally
- [ ] User-facing docs or screenshots -- no user-facing string

**End to end, through the real tool**, with the exact payload that failed five times in production:

```
files sent as:            str -> [{"path": ".../poster.png", "title": "..."}]
returned:                 Delivered 1 file(s): poster.png (72 B).
still "should be array":  False
registered in the store:  True
```

Twelve cases, five mutations, each killing at least one:

| mutation | red |
|---|---|
| drop the whole branch (the old behaviour) | 3 |
| raise instead of falling through on a parse failure | 5 |
| accept the parse whatever type it yielded | 1 |
| do not re-cast the parsed value, so items are not cast | 1 |
| widen the tuple to include `string` | **survived -- unreachable** |

The last is worth stating rather than hiding: that mutation cannot change behaviour, because the `type == "string"` branch returns earlier and the type check would refuse the result anyway. I also tried moving the new block above that branch, alone and combined with widening the tuple, and neither reddens anything either. The invariant holds by construction; the case that documents it says so.

**Pre-existing failures, checked against the base.** 7 failed in the full run, all in `tests/test_security_network.py` (`test_a_redirect_to_a_dangerous_address_is_refused` and siblings). Running that file alone on this same tree gives 4 -- the count is order-dependent -- and the change touches only `raven/agent/tools/params.py` plus a new test file, neither of which `raven/security/network.py` imports. The failures are the box's resolver: it returns poisoned answers roughly half the time, which I measured separately (`en.wikipedia.org` came back as `2001::1` on 3 of 6 lookups). Not introduced here, and worth someone's attention on its own.

## Risk

- [x] Security impact considered -- widens what reaches a tool's own validation, not what bypasses it. A parsed value is still validated against the same schema, and a parse that yields the wrong type is refused exactly as before
- [x] Backward compatibility considered -- a call that worked before takes the identical path; the new branch is reached only by a `str` where an array or object is declared, which previously always failed
- [x] Rollback path is clear -- single-commit revert

## Related Issues

N/A

## Swept

`mr-review-patterns` pre-submit sweep, over `git diff origin/main...HEAD`.

- **1 (a write location moved)** -- nothing moved; a branch was added to an existing dispatch on `target_type`. Grepped every caller of `cast_params`: the registry is the only one, and it runs `validate_params` right after, which is the reader this affects.
- **2 (a value with several entry points)** -- the cast is the single funnel every tool argument passes through, which is why the fix belongs here and not in `deliver_files`.
- **3 (a guard)** -- checked both directions: what it wrongly lets through (JSON of the wrong type, non-JSON text -- both refused, with cases) and what it wrongly refuses (a real list, unaffected, with a case).
- **4 (tests)** -- five mutations above, one honestly reported as unreachable. Failing-ID sets diffed against the base rather than counts.
- **6 (stale text)** -- the module docstring enumerated what is coerced ("a string that should be an integer"); it now names arrays and objects too. Grepped the tracked tree for other statements of the rule: none.
- **9 (error handling)** -- the new path adds one way to end and it is caught: `json.loads` raises `ValueError` on bad text and `TypeError` on a non-string, and both fall through to the untouched value.
- **11** -- `git ls-files | grep -c node_modules` is 0; the commit-message gate was run, not eyeballed.

Not applicable, with reasons: **3b, 3c, 3d, 3e** (no tool schema produced, no state moved, one lane, no assertion lost its symbol); **4b, 4c, 4d** (no platform branch, no shell, no rendered number); **5, 5b** (no dependency; `json` is stdlib and already imported across this package); **6b, 6c** (no backup, no generated artifact); **7, 7b** (nothing on disk); **8** (no asset, no secret); **10** (no prompt text).

## Deliberately not fixed

The error text still ends with `[Analyze the error above and try a different approach.]`, and the model plainly did not -- it sent the same shape five times. A message naming the expected spelling (`files must be a JSON array, not a string containing one`) would give it something to act on. That is a change to the registry's error surface, affects every tool, and deserves its own argument.

## Moved from !594

Opened on a compliant source branch. AGENTS.md 2.1 asks for a 3-5 word snake_case description and the previous branch had seven; GitLab cannot change an existing merge request's source branch, so the change moves rather than the name.

Same commits, rebased onto `c2a9768f`. Both code findings from the review on !594 are already in them -- `RecursionError` caught with its regression case, and the implementation comment reduced to the two non-obvious reasons. Nothing else changed, so that review carries.
2026-09-09 15:42:36 +00:00

201 lines
6.8 KiB
Python

"""Package skeleton smoke tests: the package imports and the paper shapes behave.
These tests pass on a fresh checkout with only Python stdlib and
pydantic + loguru installed. They do NOT require an LLM provider, a
configured workspace, or any external service.
Where the shapes live (the papers package):
TokenStrategy / UsageSnapshot raven.contracts.token_strategy
AssembledContext / TokenBudget raven.contracts.assembled
``raven.core`` is the assembly root; the tests below pin that it carries no
interface shapes of its own.
"""
from __future__ import annotations
import pytest
from raven import __version__
from raven.contracts.assembled import AssembledContext, TokenBudget
from raven.contracts.token_strategy import TokenStrategy, UsageSnapshot
# ---------------------------------------------------------------------------
# Package metadata
# ---------------------------------------------------------------------------
def test_package_imports():
assert isinstance(__version__, str)
assert __version__ # non-empty
def test_assembled_shapes_live_in_the_papers():
from raven.contracts.assembled import AssembledContext as ReAssembled
from raven.contracts.assembled import TokenBudget as ReBudget
assert ReAssembled is AssembledContext
assert ReBudget is TokenBudget
def test_raven_core_is_the_assembly_root_not_the_old_context_home():
# raven.core was once a transitional home for AssembledContext/TokenBudget
# (now in memory_engine) and was deleted with a tombstone here. The name is
# re-founded as the assembly root (the *_stack builders). Keep the old
# meaning dead: a holdout importing the context types from here must still
# fail loudly, while the assembly stacks answer at their new address.
import raven.core
assert not hasattr(raven.core, "AssembledContext")
assert not hasattr(raven.core, "TokenBudget")
from raven.core.plugin_stack import build_plugin_registry # noqa: F401
# ---------------------------------------------------------------------------
# Surviving ABC: TokenStrategy
#
# ContextEngine / Monitor / SkillHandler were removed — they had zero
# implementations and the design owners chose alternate routes
# (SkillService, Sentinel Planner, pending Curator). The TokenStrategy
# contract remains load-bearing (CacheOptimizer, UsageTracker,
# SystemAndTailCacheStrategy all implement it), so its abstractness is still
# part of the tier-1 contract.
# ---------------------------------------------------------------------------
def test_token_strategy_is_abstract():
with pytest.raises(TypeError):
TokenStrategy() # type: ignore[abstract]
def test_minimal_token_strategy_subclass():
class Noop(TokenStrategy):
@property
def name(self) -> str:
return "noop"
strat = Noop()
assert strat.name == "noop"
# ---------------------------------------------------------------------------
# Surviving dataclass behavior
# ---------------------------------------------------------------------------
def test_token_budget_is_a_plain_shape():
import dataclasses
b = TokenBudget(
context_length=100_000,
reserved_output=8_000,
reserved_tools=4_000,
reserved_system=2_000,
available_history=86_000,
)
assert dataclasses.is_dataclass(b)
assert [f.name for f in dataclasses.fields(b)] == [
"context_length",
"reserved_output",
"reserved_tools",
"reserved_system",
"available_history",
]
assert b.available_history == 86_000
def test_assembled_context_defaults():
ac = AssembledContext(messages=[{"role": "user", "content": "hi"}])
assert ac.system_prompt_addition is None
assert ac.include_indices is None
assert ac.metadata == {}
def test_usage_snapshot():
u = UsageSnapshot(
model="claude-opus-4-6",
input_tokens=10_000,
output_tokens=500,
)
# None, not 0.0: no price has been established for this call yet, and a
# plan-billed provider never establishes one.
assert u.cost_usd is None
assert u.session_key is None
# ---------------------------------------------------------------------------
# Config
# ---------------------------------------------------------------------------
def test_config_module_imports():
# Lazy import — config depends on pydantic-settings which may not load
# if the env is misconfigured. We still want the test to be informative.
from raven.config import (
ContextConfig,
RavenConfig,
SentinelConfig,
SkillForgeConfig,
TokenWiseConfig,
)
cfg = RavenConfig()
assert isinstance(cfg.context, ContextConfig)
assert isinstance(cfg.sentinel, SentinelConfig)
assert isinstance(cfg.token_wise, TokenWiseConfig)
assert isinstance(cfg.skill_forge, SkillForgeConfig)
def test_config_safe_defaults():
from raven.config import RavenConfig
cfg = RavenConfig()
# Risky/novel auto-* features must default to OFF so a fresh install
# behaves like vanilla raven; the baseline retrieval pipeline
# (context engine, skill_forge retrieval/injection) defaults ON as of R8.
assert cfg.sentinel.enabled is False
assert cfg.skill_forge.enabled is True # R8: retrieval/injection pipeline on by default
assert cfg.skill_forge.auto_detect is False
assert cfg.skill_forge.auto_evolve is False
# Baseline memory/skill feature layer defaults ON: a
# fresh install runs the everos memory backend, the SkillForgeRouter, and
# empty-response recovery. Pinned so a future silent flip gets caught.
assert cfg.memory.backend == "everos"
assert cfg.skill_forge.router.enabled is True
assert cfg.base.agents.defaults.empty_recovery_enabled is True
# Safe/cheap defaults can be ON.
assert cfg.token_wise.usage_tracking is True
assert cfg.token_wise.cache_optimization is True
def test_config_camel_and_snake_keys():
from raven.config import SentinelConfig
# snake_case
s1 = SentinelConfig(idle_threshold_seconds=600)
# camelCase (as Pydantic parses from YAML/JSON)
s2 = SentinelConfig.model_validate({"idleThresholdSeconds": 600})
assert s1.idle_threshold_seconds == 600
assert s2.idle_threshold_seconds == 600
def test_tick_interval_seconds_rejects_sub_minute_values():
"""Pydantic Field(ge=60) blocks foot-guns like 1-second ticks that
would burn Planner LLM budget faster than new inbound arrives."""
from pydantic import ValidationError
from raven.config import SentinelConfig
SentinelConfig(tick_interval_seconds=60) # boundary value OK
with pytest.raises(ValidationError):
SentinelConfig(tick_interval_seconds=30)
with pytest.raises(ValidationError):
SentinelConfig(tick_interval_seconds=1)
if __name__ == "__main__":
# Allow `python tests/test_package_skeleton.py` as a quick smoke run.
pytest.main([__file__, "-v"])