dev/nisse
2
1 Fork 0
Code Issues 23 Pull Requests Actions Packages Projects Releases 1 Wiki Activity

Move configuration to YAML, and make the tuning knobs configurable #43

Merged
dongar merged 4 commits from yaml-config into main 2026-09-23 23:49:44 +00:00
Conversation 5 Commits 4 Files Changed 35
+571 -777
28 changed files with 571 additions and 777 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit 15d69799bf - Show all commits

No files matched your search

.env.example
+12 -58
View File
@@ -1,64 +1,18 @@
# Copy to .env (chmod 600). Real environment variables win over this file.
# .env is gitignored; never commit real values.
# Secrets only. Copy to .env (chmod 600); it is gitignored, so never commit
# real values. Real environment variables win over this file.
#
# Settings can also live in a YAML file (nisse-dev.example.yaml), which keeps
# structure and the tuning knobs; secrets stay here and are referenced from
# the file as ${NISSE_LLM_TOKEN} and so on. Every variable below overrides the
# matching key in that file.
# Everything else lives in the YAML config (see nisse-dev.example.yaml),
# which refers to these by name, e.g. `token: ${NISSE_LLM_TOKEN}`. The live
# integration tests use the same names, from tests/integration/nisse-test.yaml.
dongar commented 2026-09-23 23:36:59 +00:00
Review
Copy link
Copy source

I feel like this should include some examples of the actual secrets we expect that match the example config file. Something like:

NISSE_MATRIX_TOKEN=replace-me
NISSE_GITEA_TOKEN=replace-me
NISSE_LLM_TOKEN=replace-me

I feel like this should include some examples of the actual secrets we expect that match the example config file. Something like: NISSE_MATRIX_TOKEN=replace-me NISSE_GITEA_TOKEN=replace-me NISSE_LLM_TOKEN=replace-me
dev-agent commented 2026-09-23 23:45:02 +00:00
Review
Copy link
Copy source

They're in the file, just below this comment block. The three lines didn't change in this PR, so the diff folded them away. To make the match with the example config explicit, each one now notes which key refers to it (llm.token, matrix.token, gitea.token), and the values are replace-me as you suggested. Done in e1034a2c.

They're in the file, just below this comment block. The three lines didn't change in this PR, so the diff folded them away. To make the match with the example config explicit, each one now notes which key refers to it (`llm.token`, `matrix.token`, `gitea.token`), and the values are `replace-me` as you suggested. Done in e1034a2c.
# --- LLM access (A5): the LiteLLM proxy is the single LLM destination ---
NISSE_LLM_BASE_URL=http://litellm.example:4000
NISSE_LLM_TOKEN=sk-your-litellm-virtual-key
# --- Role mapping (A11): roles resolve to concrete model strings ---
NISSE_ROLE_EXECUTOR=anthropic/claude-haiku-4-5-20251001
NISSE_ROLE_PLANNER=anthropic/claude-opus-5
NISSE_ROLE_DELEGATE=anthropic/claude-opus-5
NISSE_ROLE_CONSULTANT=anthropic/claude-opus-5
# Opt-in for the live planning test: a model the virtual key can reach that
# holds the tool-call contract. Planning escalates unconditionally (A12), and
# small local models do not hold the contract reliably — llama3.2 returns
# truncated tool arguments, Haiku-class and up are solid.
NISSE_LIVE_PLANNER_MODEL=claude-haiku-4-5
# --- Matrix transport (A1), used by nisse-dev's MatrixTransport ---
NISSE_MATRIX_HOMESERVER=https://matrix.example
NISSE_MATRIX_USER_ID=@nisse-dev:example
NISSE_MATRIX_TOKEN=syt_your_matrix_access_token
NISSE_MATRIX_ROOM_ID=!roomid:example
# Dedicated room for the live integration test; never the production room.
NISSE_MATRIX_TEST_ROOM_ID=!testroomid:example
NISSE_MATRIX_OPERATORS=@you:example
# --- Gitea (B2): service account for branches, PRs, and comment intake ---
NISSE_GITEA_URL=https://gitea.example
NISSE_GITEA_TOKEN=your-scoped-service-account-token
# Sandbox repo for live tests; never the real project repo.
NISSE_GITEA_TEST_REPO=owner/sandbox-repo
# Gitea logins allowed to drive the agent via PR comments (A4 per-channel
# allowlist; a Matrix MXID and a Gitea login are different names).
NISSE_GITEA_OPERATORS=your-gitea-login
# --- Running the agent (the demo, not the tests) ---
# The tests use NISSE_GITEA_TEST_REPO; a running agent uses these.
#
# Layout. NISSE_WORKSPACE is the *clone the agent works in*, not the directory
# the agent runs from — they must not be the same place:
#
# ~/nisse-demo-workspace/ deployment root, not a git repo
# ├── .env this file
# ├── state.db the A2 state store
# └── project/ NISSE_WORKSPACE, the clone
#
# The agent commits its workspace with `git add -A` and pushes it, so anything
# inside the clone reaches the project repository. A .env or state store placed
# there would publish credentials as a side effect of the agent working
# correctly; startup refuses to proceed if either is inside the workspace.
NISSE_STATE_DB=~/nisse-demo-workspace/state.db
NISSE_WORKSPACE=~/nisse-demo-workspace/project
NISSE_PROJECT=demo
NISSE_GITEA_REPO=owner/demo-repo
# NISSE_BASE_BRANCH=main
# Path to this file, if not ./.env
# NISSE_ENV=~/nisse-demo-workspace/.env
# Where the config file is, if not ./nisse-dev.yaml. Where this file is, if
# not ./.env, is NISSE_ENV, which has to be set in the real environment.
# NISSE_CONFIG=~/nisse-demo-workspace/nisse-dev.yaml
# Opt-in for the capability matrix: models to measure, comma separated.
# Spends real tokens across every model listed. See docs/Tuning-Guide.md.
# NISSE_CAPABILITY_MODELS=local-fast,local-coder
.gitea/workflows/ci.yml
+3 -11
View File
@@ -34,19 +34,11 @@ jobs:
run: uv run pytest -m "not integration"
- name: Integration tests
# Live-path credentials come from repo Actions secrets; the tests
# skip-guard on these, so a missing secret or unreachable service
# degrades to "skipped", never a failure on unrelated work.
# Everything but the secrets is in tests/integration/nisse-test.yaml,
# which refers to these by name. A missing secret fails the live
# tests that need it; it is never quietly turned into a skip.
env:
NISSE_LLM_BASE_URL: ${{ secrets.NISSE_LLM_BASE_URL }}
NISSE_LLM_TOKEN: ${{ secrets.NISSE_LLM_TOKEN }}
NISSE_ROLE_EXECUTOR: ${{ secrets.NISSE_ROLE_EXECUTOR }}
NISSE_LIVE_PLANNER_MODEL: ${{ secrets.NISSE_LIVE_PLANNER_MODEL }}
NISSE_MATRIX_HOMESERVER: ${{ secrets.NISSE_MATRIX_HOMESERVER }}
NISSE_MATRIX_USER_ID: ${{ secrets.NISSE_MATRIX_USER_ID }}
NISSE_MATRIX_TOKEN: ${{ secrets.NISSE_MATRIX_TOKEN }}
NISSE_MATRIX_TEST_ROOM_ID: ${{ secrets.NISSE_MATRIX_TEST_ROOM_ID }}
NISSE_GITEA_URL: ${{ secrets.NISSE_GITEA_URL }}
NISSE_GITEA_TOKEN: ${{ secrets.NISSE_GITEA_TOKEN }}
NISSE_GITEA_TEST_REPO: ${{ secrets.NISSE_GITEA_TEST_REPO }}
run: uv run pytest -m integration -rs
docs/Architecture-Build-Spec.md
+4 -2
View File
@@ -389,8 +389,10 @@ Requirements:
- **No vendor credentials on the agent VM.** LiteLLM holds every key, sourced from
Bitwarden machine accounts (A7); the agent holds one LiteLLM token. Adding a vendor
does not touch the agent VM, and a compromised agent leaks no provider keys.
- Config from env/`.env`, **nothing hardcoded**: base URL, auth token, and the role
mapping (A11). Changing any of it is an env change + restart, nothing more.
- Config from the agent's config file, **nothing hardcoded**: base URL, auth token,
and the role mapping (A11). Secrets in it are references to environment
variables, never values. Changing any of it is a config edit + restart,
nothing more.
- **Budget caps and spend alerts live on the LiteLLM virtual key**, not in agent code.
### A6. Browser Tool
docs/Tuning-Guide.md
-1
View File
@@ -131,7 +131,6 @@ role.
executor. Frontier planning wants the large default. This one knob covers
every role for now; a per-model setting is coming with the context budget
(see *Coming next*).
- Environment override: `NISSE_MAX_OUTPUT_TOKENS`.
### `tuning.executor.max_turns` — default 24
nisse-dev.example.yaml
+18 -20
View File
@@ -2,44 +2,42 @@
# NISSE_CONFIG at your copy) and edit.
#
# - Structure lives here. Secrets do not: write them as ${VARIABLE} references,
# resolved from the environment (or .env) at startup.
# - Environment variables win. Every NISSE_* variable in .env.example
# overrides the key noted beside it here.
# resolved from the environment (or .env, see .env.example) at startup.
# - Keep this file outside the workspace clone. Startup refuses otherwise,
# because the agent commits and pushes everything in the clone.
# - Unknown keys are errors, so a misspelt knob fails at startup instead of
# silently doing nothing.
state_db: ~/nisse-demo-workspace/state.db # NISSE_STATE_DB
workspace: ~/nisse-demo-workspace/project # NISSE_WORKSPACE, the clone
project: demo # NISSE_PROJECT
state_db: ~/nisse-demo-workspace/state.db
workspace: ~/nisse-demo-workspace/project # the clone
project: demo
llm: # A5: LiteLLM is the only LLM destination
base_url: http://litellm.example:4000 # NISSE_LLM_BASE_URL
base_url: http://litellm.example:4000
token: ${NISSE_LLM_TOKEN}
# Ceiling on each response. Must fit inside a local model's context window
# alongside the prompt; see docs/Tuning-Guide.md.
max_output_tokens: 16384 # NISSE_MAX_OUTPUT_TOKENS
max_output_tokens: 16384
roles: # A11: role -> LiteLLM model alias
executor: local-fast # NISSE_ROLE_EXECUTOR
planner: claude-opus-5 # NISSE_ROLE_PLANNER
delegate: claude-opus-5 # NISSE_ROLE_DELEGATE
consultant: claude-opus-5 # NISSE_ROLE_CONSULTANT
executor: local-fast
planner: claude-opus-5
delegate: claude-opus-5
consultant: claude-opus-5
matrix: # A1
homeserver: https://matrix.example # NISSE_MATRIX_HOMESERVER
user_id: "@nisse-dev:example" # NISSE_MATRIX_USER_ID
homeserver: https://matrix.example
user_id: "@nisse-dev:example"
token: ${NISSE_MATRIX_TOKEN}
room_id: "!roomid:example" # NISSE_MATRIX_ROOM_ID
operators: ["@you:example"] # NISSE_MATRIX_OPERATORS (comma-separated)
room_id: "!roomid:example"
operators: ["@you:example"]
gitea: # B2
url: https://gitea.example # NISSE_GITEA_URL
url: https://gitea.example
token: ${NISSE_GITEA_TOKEN}
repo: owner/demo-repo # NISSE_GITEA_REPO
operators: [your-gitea-login] # NISSE_GITEA_OPERATORS (comma-separated)
base_branch: main # NISSE_BASE_BRANCH
repo: owner/demo-repo
operators: [your-gitea-login]
base_branch: main
# What each knob does, and how to choose a value: docs/Tuning-Guide.md.
# These are the defaults; the section can be left out entirely. The file is
nisse-dev/src/nisse_dev/catalog.py
+2 -2
View File
@@ -121,8 +121,8 @@ class ModelCatalog:
Reported together rather than one per boot: an operator fixing a
config wants the whole list, not to rediscover the next typo on the
next start (the same reason ``AgentConfig.from_env`` batches its
missing variables).
next start (the same reason ``AgentConfig`` reports every missing
setting at once).
"""
unknown = {
role: alias for role, alias in roles.items() if alias not in self._facts
nisse-dev/src/nisse_dev/config.py
+37 -68
View File
@@ -1,14 +1,11 @@
"""Agent configuration (spec A5: nothing hardcoded).
Configuration comes from a YAML file plus the environment:
- **The file carries structure.** Roles, connections, and the tuning knobs
(see docs/Tuning-Guide.md). It holds no secrets: a credential is written as
a reference, ``token: ${NISSE_LLM_TOKEN}``, and resolved from the
environment, so the file can be reviewed or shared without leaking.
- **The environment wins.** Every setting that existed as an environment
variable still does, and overrides the file. The file is optional, so a
deployment or CI run configured purely from the environment works as before.
Configuration is a YAML file: roles, connections, and the tuning knobs (see
docs/Tuning-Guide.md). It holds no secrets. A credential is written as a
reference, ``token: ${NISSE_LLM_TOKEN}``, and resolved from the environment,
so the file can be reviewed, shared, or committed without leaking. The
environment supplies those references and nothing else: every setting has
exactly one home.
Every value is collected and validated **before** anything is built, and a
single error names every problem at once. The alternative — failing at first
@@ -31,32 +28,6 @@ from nisse_dev.matrix import MatrixConfig
DEFAULT_CONFIG_FILE = "nisse-dev.yaml"
# Each environment variable and the key it overrides. The names are the ones
# the agent has always used, so existing .env files and CI keep working.
ENV_KEYS: dict[str, str] = {
"NISSE_STATE_DB": "state_db",
"NISSE_WORKSPACE": "workspace",
"NISSE_PROJECT": "project",
"NISSE_LLM_BASE_URL": "llm.base_url",
"NISSE_LLM_TOKEN": "llm.token",
"NISSE_MAX_OUTPUT_TOKENS": "llm.max_output_tokens",
"NISSE_ROLE_EXECUTOR": "roles.executor",
"NISSE_ROLE_PLANNER": "roles.planner",
"NISSE_ROLE_DELEGATE": "roles.delegate",
"NISSE_ROLE_CONSULTANT": "roles.consultant",
"NISSE_MATRIX_HOMESERVER": "matrix.homeserver",
"NISSE_MATRIX_USER_ID": "matrix.user_id",
"NISSE_MATRIX_TOKEN": "matrix.token",
"NISSE_MATRIX_ROOM_ID": "matrix.room_id",
"NISSE_MATRIX_OPERATORS": "matrix.operators",
"NISSE_GITEA_URL": "gitea.url",
"NISSE_GITEA_TOKEN": "gitea.token",
"NISSE_GITEA_REPO": "gitea.repo",
"NISSE_GITEA_OPERATORS": "gitea.operators",
"NISSE_BASE_BRANCH": "gitea.base_branch",
}
KEY_ENV = {key: env for env, key in ENV_KEYS.items()}
REQUIRED_KEYS = (
"state_db",
"workspace",
@@ -78,8 +49,6 @@ REQUIRED_KEYS = (
"gitea.operators",
)
OPTIONAL_KEYS = ("llm.max_output_tokens", "gitea.base_branch")
# Kept for callers that list what a pure-environment deployment must set.
REQUIRED = tuple(KEY_ENV[key] for key in REQUIRED_KEYS)
_REFERENCE = re.compile(r"\$\{([A-Za-z_][A-Za-z0-9_]*)\}")
@@ -226,11 +195,6 @@ def _read_file(path: Path, errors: list[str]) -> dict[str, Any]:
return loaded
def _describe(key: str) -> str:
env = KEY_ENV.get(key)
return f"{key} ({env})" if env else key
@dataclass(frozen=True)
class AgentConfig:
state_db: Path
@@ -252,28 +216,39 @@ class AgentConfig:
@classmethod
def load(cls, env: Mapping[str, str] = os.environ) -> "AgentConfig":
"""The file named by NISSE_CONFIG, or ./nisse-dev.yaml if present,
overlaid by the environment. Naming a file that does not exist is an
error; the default file is simply optional."""
named = env.get("NISSE_CONFIG")
path = Path(named or DEFAULT_CONFIG_FILE).expanduser()
if named and not path.is_file():
"""The file named by NISSE_CONFIG, or ./nisse-dev.yaml."""
path = Path(env.get("NISSE_CONFIG") or DEFAULT_CONFIG_FILE).expanduser()
if not path.is_file():
raise ConfigError(
f"cannot start: NISSE_CONFIG names {path}, which is missing"
f"cannot start: no config file at {path}; create it from "
"nisse-dev.example.yaml, or point NISSE_CONFIG at one"
)
return cls.from_sources(env, path if path.is_file() else None)
return cls.from_file(path, env)
@classmethod
def from_env(cls, env: Mapping[str, str] = os.environ) -> "AgentConfig":
"""Environment only, with no file."""
return cls.from_sources(env, None)
@classmethod
def from_sources(
cls, env: Mapping[str, str], config_file: Path | None
def from_file(
cls, path: Path, env: Mapping[str, str] = os.environ
) -> "AgentConfig":
errors: list[str] = []
tree = _read_file(config_file, errors) if config_file else {}
tree = _read_file(path, errors)
return cls._build(tree, env, errors, config_file=path)
@classmethod
def from_tree(
cls, tree: Mapping[str, Any], env: Mapping[str, str] = os.environ
) -> "AgentConfig":
"""From an already-parsed config, as the file would give it."""
return cls._build(tree, env, [], config_file=None)
@classmethod
def _build(
cls,
tree: Mapping[str, Any],
env: Mapping[str, str],
errors: list[str],
*,
config_file: Path | None,
) -> "AgentConfig":
tuning = _parse_tuning(tree.get("tuning"), errors)
flat = _flatten(tree)
flat.pop("tuning", None)
@@ -281,20 +256,16 @@ class AgentConfig:
known = set(REQUIRED_KEYS) | set(OPTIONAL_KEYS)
for key in sorted(set(flat) - known):
errors.append(f"{key}: unknown setting")
overrides = {key: env[var] for var, key in ENV_KEYS.items() if var in env}
# A key the environment overrides is not resolved from the file, so a
# reference there to an unset variable is not an error.
values = {
key: _resolve(value, key, env, errors)
for key, value in flat.items()
if key in known and key not in overrides
if key in known
}
values.update(overrides)
# Whitespace-only counts as missing: a token of spaces would pass a
# truthiness check and fail at the first request instead.
missing = [
_describe(key)
key
for key in REQUIRED_KEYS
if values.get(key) is None
or (isinstance(values[key], str) and not values[key].strip())
@@ -309,9 +280,7 @@ class AgentConfig:
if max_output <= 0:
raise ValueError
except (TypeError, ValueError):
errors.append(
f"{_describe('llm.max_output_tokens')}: must be a positive integer"
)
errors.append("llm.max_output_tokens: must be a positive integer")
if errors:
raise ConfigError("cannot start: " + "; ".join(errors))
@@ -357,7 +326,7 @@ class AgentConfig:
"""
workspace = self.workspace.expanduser().resolve()
candidates = [
("NISSE_STATE_DB", self.state_db),
("state_db", self.state_db),
("NISSE_ENV", Path(env.get("NISSE_ENV", ".env"))),
]
if self.config_file is not None:
nisse-dev/src/nisse_dev/gitea.py
-10
View File
@@ -10,8 +10,6 @@ idempotent across a crash: boot reconciliation (A3) asks Gitea whether the
effect already happened rather than repeating it.
"""
import os
from collections.abc import Mapping
from dataclasses import dataclass
from typing import Any
@@ -60,14 +58,6 @@ class GiteaClient:
transport=transport,
)
@classmethod
def from_env(cls, env: Mapping[str, str] = os.environ) -> "GiteaClient":
base_url = env.get("NISSE_GITEA_URL")
token = env.get("NISSE_GITEA_TOKEN")
if not base_url or not token:
raise GiteaError("set NISSE_GITEA_URL and NISSE_GITEA_TOKEN")
return cls(base_url, token)
def _request(self, method: str, path: str, **kwargs: Any) -> Any:
try:
response = self._http.request(method, path, **kwargs)
nisse-dev/src/nisse_dev/llm.py
+9 -46
View File
@@ -2,18 +2,16 @@
One client interface, present from the first real LLM call. All traffic goes
to the LiteLLM proxy — the VM's single LLM destination; the agent holds one
LiteLLM token and no vendor credentials. Config comes from the environment
(``NISSE_LLM_BASE_URL``, ``NISSE_LLM_TOKEN``), nothing hardcoded.
LiteLLM token and no vendor credentials. The base URL and token come from the
agent's config (nisse_dev.config), nothing hardcoded.
Model identity is expressed as roles (A11), resolved to concrete model strings
from env config; the resolved mapping is pinned at task start so a mid-task
from config; the resolved mapping is pinned at task start so a mid-task
config change cannot split a task across two planners.
"""
import json
import os
import time
from collections.abc import Mapping
from dataclasses import asdict, dataclass
from typing import Any, ClassVar, Protocol
@@ -49,11 +47,6 @@ RATE_LIMIT_RETRY_S = 20.0
DEFAULT_MAX_OUTPUT_TOKENS = 16384
def max_output_tokens(env: Mapping[str, str] = os.environ) -> int:
raw = env.get("NISSE_MAX_OUTPUT_TOKENS")
return int(raw) if raw else DEFAULT_MAX_OUTPUT_TOKENS
class UpstreamUnavailable(LLMError):
"""The proxy or provider could not be reached, or refused for reasons
unrelated to what was asked: a network blip, a timeout, a 5xx, a rate
@@ -113,22 +106,10 @@ class RoleMap:
delegate: str
consultant: str
_ENV_VARS: ClassVar[dict[str, str]] = {
"executor": "NISSE_ROLE_EXECUTOR",
"planner": "NISSE_ROLE_PLANNER",
"delegate": "NISSE_ROLE_DELEGATE",
"consultant": "NISSE_ROLE_CONSULTANT",
}
@classmethod
def from_env(cls, env: Mapping[str, str] = os.environ) -> "RoleMap":
missing = [var for var in cls._ENV_VARS.values() if not env.get(var)]
if missing:
raise LLMError(f"role mapping incomplete; set: {', '.join(missing)}")
return cls(**{role: env[var] for role, var in cls._ENV_VARS.items()})
ROLES: ClassVar[tuple[str, ...]] = ("executor", "planner", "delegate", "consultant")
def resolve(self, role: str) -> str:
if role not in self._ENV_VARS:
if role not in self.ROLES:
raise LLMError(f"unknown role: {role!r}")
return str(getattr(self, role))
@@ -156,20 +137,12 @@ class LiteLLMClient:
)
self._renderer = OpenAIChatRenderer()
self._sleep = sleep
self._max_tokens = max_tokens or max_output_tokens()
self._max_tokens = max_tokens or DEFAULT_MAX_OUTPUT_TOKENS
# Set after each call so a caller can meter itself (A13 caps);
# a side channel rather than a return value, so the LLMClient
# protocol stays the same for every implementation.
self.last_usage: dict[str, int] = {}
@classmethod
def from_env(cls, env: Mapping[str, str] = os.environ) -> "LiteLLMClient":
base_url = env.get("NISSE_LLM_BASE_URL")
token = env.get("NISSE_LLM_TOKEN")
if not base_url or not token:
raise LLMError("set NISSE_LLM_BASE_URL and NISSE_LLM_TOKEN")
return cls(base_url, token)
def complete(
self,
events: list[StoredEvent],
@@ -203,7 +176,7 @@ class LiteLLMClient:
if choice.get("finish_reason") == "length":
raise TruncatedResponse(
f"{requested_model} hit its output limit mid-answer; "
"raise NISSE_MAX_OUTPUT_TOKENS"
"raise llm.max_output_tokens"
)
produced_by = data.get("model") or requested_model
usage = data.get("usage") or {}
@@ -275,20 +248,10 @@ class AnthropicPassthroughClient:
transport=transport,
)
self._renderer = AnthropicMessagesRenderer()
self._max_tokens = max_tokens or max_output_tokens()
self._max_tokens = max_tokens or DEFAULT_MAX_OUTPUT_TOKENS
self._sleep = sleep
self.last_usage: dict[str, int] = {}
@classmethod
def from_env(
cls, env: Mapping[str, str] = os.environ
) -> "AnthropicPassthroughClient":
base_url = env.get("NISSE_LLM_BASE_URL")
token = env.get("NISSE_LLM_TOKEN")
if not base_url or not token:
raise LLMError("set NISSE_LLM_BASE_URL and NISSE_LLM_TOKEN")
return cls(base_url, token)
def complete(
self,
events: list[StoredEvent],
@@ -315,7 +278,7 @@ class AnthropicPassthroughClient:
if data.get("stop_reason") == "max_tokens":
raise TruncatedResponse(
f"{requested_model} hit the output limit of {self._max_tokens} "
f"tokens mid-answer; raise NISSE_MAX_OUTPUT_TOKENS"
f"tokens mid-answer; raise llm.max_output_tokens"
)
content = data.get("content")
if not isinstance(content, list):
nisse-dev/src/nisse_dev/matrix.py
+1 -26
View File
@@ -11,12 +11,10 @@ loop backs off (capped) and retries indefinitely — an outage delays reactions
but never fails anything.
"""
import os
import threading
import uuid
from collections.abc import Callable, Mapping
from collections.abc import Callable
from dataclasses import dataclass
from typing import ClassVar
import httpx
@@ -34,29 +32,6 @@ class MatrixConfig:
token: str
room_id: str
_ENV: ClassVar[tuple[str, ...]] = (
"NISSE_MATRIX_HOMESERVER",
"NISSE_MATRIX_USER_ID",
"NISSE_MATRIX_TOKEN",
)
@classmethod
def from_env(
cls,
env: Mapping[str, str] = os.environ,
room_var: str = "NISSE_MATRIX_ROOM_ID",
) -> "MatrixConfig":
needed = (*cls._ENV, room_var)
missing = [var for var in needed if not env.get(var)]
if missing:
raise TransportError(f"matrix config incomplete; set: {', '.join(missing)}")
return cls(
homeserver=env["NISSE_MATRIX_HOMESERVER"].rstrip("/"),
user_id=env["NISSE_MATRIX_USER_ID"],
token=env["NISSE_MATRIX_TOKEN"],
room_id=env[room_var],
)
class MatrixTransport:
def __init__(
nisse-dev/tests/test_agent_build.py
+13 -12
View File
@@ -8,7 +8,7 @@ from typing import Any
import httpx
import pytest
from test_config import COMPLETE
from test_config import SECRETS, tree
from nisse_dev.agent import build_agent, discover_models, prepare_workspace
from nisse_dev.catalog import ModelCatalog, UnknownModel
@@ -34,14 +34,15 @@ TEST_CATALOG = ModelCatalog.from_payload(
)
def config_for(tmp_path: Path, **overrides: str) -> AgentConfig:
env = {
**COMPLETE,
"NISSE_STATE_DB": str(tmp_path / "state.db"),
"NISSE_WORKSPACE": str(tmp_path / "work"),
**overrides,
}
return AgentConfig.from_env(env)
def config_for(tmp_path: Path, **changes: str) -> AgentConfig:
return AgentConfig.from_tree(
tree(
state_db=str(tmp_path / "state.db"),
workspace=str(tmp_path / "work"),
**changes,
),
SECRETS,
)
@pytest.fixture
@@ -59,7 +60,7 @@ def origin(tmp_path: Path) -> Path:
def test_workspace_is_cloned_when_absent(tmp_path: Path, origin: Path) -> None:
config = config_for(tmp_path, NISSE_GITEA_URL=f"file://{origin}")
config = config_for(tmp_path, gitea__url=f"file://{origin}")
# authenticated_url would mangle a file:// remote, so clone directly.
GitRepo(tmp_path).clone(str(origin), config.workspace)
repo = prepare_workspace(config, "nisse-dev-testing")
@@ -206,7 +207,7 @@ def test_httpx_clients_are_not_shared_between_routes(
def test_a_typod_role_refuses_to_boot(tmp_path: Path, monkeypatch) -> None:
# Misconfiguration should cost a second at startup, not be discovered ten
# minutes into a milestone.
config = config_for(tmp_path, NISSE_ROLE_PLANNER="clause-opus-5")
config = config_for(tmp_path, roles__planner="clause-opus-5")
monkeypatch.setattr(
"nisse_dev.agent.ModelCatalog.discover",
classmethod(lambda cls, *a, **kw: TEST_CATALOG),
@@ -220,7 +221,7 @@ def test_a_role_this_key_cannot_reach_refuses_to_boot(
) -> None:
# /model/info is scoped per key, so the same check catches a role pointed
# at a real model that this token has no access to.
config = config_for(tmp_path, NISSE_ROLE_EXECUTOR="local-fast")
config = config_for(tmp_path, roles__executor="local-fast")
monkeypatch.setattr(
"nisse_dev.agent.ModelCatalog.discover",
classmethod(lambda cls, *a, **kw: TEST_CATALOG),
nisse-dev/tests/test_config.py
+225 -235
View File
@@ -1,163 +1,37 @@
import copy
from pathlib import Path
from typing import Any
import pytest
from nisse_dev.config import AgentConfig, ConfigError
COMPLETE = {
"NISSE_STATE_DB": "~/demo/state.db",
"NISSE_WORKSPACE": "~/demo/work",
"NISSE_PROJECT": "greet",
"NISSE_LLM_BASE_URL": "https://litellm.test",
"NISSE_LLM_TOKEN": "sk-x",
"NISSE_ROLE_EXECUTOR": "claude-haiku-4-5",
"NISSE_ROLE_PLANNER": "claude-opus-5",
"NISSE_ROLE_DELEGATE": "claude-opus-5",
"NISSE_ROLE_CONSULTANT": "claude-haiku-4-5",
"NISSE_MATRIX_HOMESERVER": "https://matrix.test",
"NISSE_MATRIX_USER_ID": "@nisse-dev-testing:test",
"NISSE_MATRIX_TOKEN": "syt_x",
"NISSE_MATRIX_ROOM_ID": "!room:test",
"NISSE_MATRIX_OPERATORS": "@don:test",
"NISSE_GITEA_URL": "https://gitea.test",
"NISSE_GITEA_TOKEN": "gt",
"NISSE_GITEA_REPO": "don/demo",
"NISSE_GITEA_OPERATORS": "don",
# A complete config as the YAML file would give it, with secrets by reference.
COMPLETE: dict[str, Any] = {
"state_db": "~/demo/state.db",
"workspace": "~/demo/work",
"project": "greet",
"llm": {"base_url": "https://litellm.test", "token": "${TEST_LLM_TOKEN}"},
"roles": {
"executor": "claude-haiku-4-5",
"planner": "claude-opus-5",
"delegate": "claude-opus-5",
"consultant": "claude-haiku-4-5",
},
"matrix": {
"homeserver": "https://matrix.test/",
"user_id": "@nisse-dev-testing:test",
"token": "${TEST_MATRIX_TOKEN}",
"room_id": "!room:test",
"operators": ["@don:test"],
},
"gitea": {
"url": "https://gitea.test",
"token": "${TEST_GITEA_TOKEN}",
"repo": "don/demo",
"operators": ["don"],
},
}
def test_complete_environment_builds() -> None:
config = AgentConfig.from_env(COMPLETE)
assert config.project == "greet"
assert config.roles.planner == "claude-opus-5"
assert config.matrix.room_id == "!room:test"
assert config.gitea_repo == "don/demo"
assert config.base_branch == "main"
def test_paths_expand_the_home_directory() -> None:
config = AgentConfig.from_env(COMPLETE)
assert config.workspace.is_absolute()
assert "~" not in str(config.state_db)
def test_every_missing_variable_is_reported_at_once() -> None:
# Failing at first use would surface a missing token twenty minutes into
# a milestone, on the call that mattered.
partial = {k: v for k, v in COMPLETE.items() if k != "NISSE_MATRIX_TOKEN"}
partial.pop("NISSE_GITEA_REPO")
with pytest.raises(ConfigError) as excinfo:
AgentConfig.from_env(partial)
message = str(excinfo.value)
assert "NISSE_MATRIX_TOKEN" in message
assert "NISSE_GITEA_REPO" in message
def test_empty_is_treated_as_missing() -> None:
with pytest.raises(ConfigError, match="NISSE_GITEA_TOKEN"):
AgentConfig.from_env({**COMPLETE, "NISSE_GITEA_TOKEN": " "})
def test_operator_lists_are_split_and_trimmed() -> None:
config = AgentConfig.from_env(
{
**COMPLETE,
"NISSE_MATRIX_OPERATORS": " @don:test , @other:test ,",
"NISSE_GITEA_OPERATORS": "don,",
}
)
assert config.operators == {"@don:test", "@other:test"}
# A trailing comma must not admit an empty identity into an allowlist.
assert "" not in config.operators
assert config.gitea_operators == {"don"}
def test_base_branch_is_overridable() -> None:
config = AgentConfig.from_env({**COMPLETE, "NISSE_BASE_BRANCH": "trunk"})
assert config.base_branch == "trunk"
def test_redacted_summary_carries_no_secrets() -> None:
summary = AgentConfig.from_env(COMPLETE).redacted()
flat = " ".join(summary.values())
for secret in ("sk-x", "syt_x", "gt"):
assert secret not in flat
assert summary["project"] == "greet"
def test_state_db_and_workspace_are_separate(tmp_path: Path) -> None:
# A2: build artefacts must not share a volume with agent state.
config = AgentConfig.from_env(COMPLETE)
assert config.state_db.parent != config.workspace
def test_state_db_inside_the_clone_is_refused(tmp_path: Path) -> None:
# git add -A would commit it, and the push would publish the agent's
# own state alongside the project.
work = tmp_path / "project"
with pytest.raises(ConfigError, match="inside the workspace"):
AgentConfig.from_env(
{
**COMPLETE,
"NISSE_WORKSPACE": str(work),
"NISSE_STATE_DB": str(work / "state.db"),
}
)
def test_env_file_inside_the_clone_is_refused(tmp_path: Path) -> None:
# The severe case: the agent would commit and push its own credentials
# as a side effect of working correctly.
work = tmp_path / "project"
with pytest.raises(ConfigError, match="NISSE_ENV"):
AgentConfig.from_env(
{
**COMPLETE,
"NISSE_WORKSPACE": str(work),
"NISSE_STATE_DB": str(tmp_path / "state.db"),
"NISSE_ENV": str(work / ".env"),
}
)
def test_sibling_paths_are_fine(tmp_path: Path) -> None:
config = AgentConfig.from_env(
{
**COMPLETE,
"NISSE_WORKSPACE": str(tmp_path / "project"),
"NISSE_STATE_DB": str(tmp_path / "state.db"),
"NISSE_ENV": str(tmp_path / ".env"),
}
)
assert config.workspace.name == "project"
# --- the YAML file (#32) ---------------------------------------------------
FILE = """
state_db: ~/demo/state.db
workspace: ~/demo/work
project: greet
llm:
base_url: https://litellm.test
token: ${TEST_LLM_TOKEN}
roles:
executor: local-fast
planner: claude-opus-5
delegate: claude-opus-5
consultant: claude-haiku-4-5
matrix:
homeserver: https://matrix.test/
user_id: "@nisse-dev-testing:test"
token: ${TEST_MATRIX_TOKEN}
room_id: "!room:test"
operators: ["@don:test", " "]
gitea:
url: https://gitea.test
token: ${TEST_GITEA_TOKEN}
repo: don/demo
operators: [don]
"""
SECRETS = {
"TEST_LLM_TOKEN": "sk-x",
"TEST_MATRIX_TOKEN": "syt_x",
@@ -165,86 +39,133 @@ SECRETS = {
}
def write(tmp_path: Path, text: str) -> Path:
path = tmp_path / "nisse-dev.yaml"
path.write_text(text)
return path
def tree(**changes: Any) -> dict[str, Any]:
"""COMPLETE with dotted keys changed; a value of None removes the key."""
result = copy.deepcopy(COMPLETE)
for dotted, value in changes.items():
*parents, leaf = dotted.split("__")
node = result
for part in parents:
node = node.setdefault(part, {})
if value is None:
node.pop(leaf, None)
else:
node[leaf] = value
return result
def test_a_file_configures_everything_with_secrets_by_reference(
tmp_path: Path,
) -> None:
config = AgentConfig.from_sources(SECRETS, write(tmp_path, FILE))
def build(**changes: Any) -> AgentConfig:
return AgentConfig.from_tree(tree(**changes), SECRETS)
def test_complete_config_builds_with_secrets_by_reference() -> None:
config = build()
assert config.project == "greet"
assert config.llm_token == "sk-x"
assert config.gitea_token == "gt"
assert config.roles.executor == "local-fast"
assert config.operators == {"@don:test"} # a blank entry admits no one
assert config.roles.planner == "claude-opus-5"
assert config.matrix.room_id == "!room:test"
assert config.matrix.homeserver == "https://matrix.test"
assert config.gitea_repo == "don/demo"
assert config.base_branch == "main"
assert config.tuning.executor.max_turns == 24 # absent section: defaults
def test_the_environment_wins_over_the_file(tmp_path: Path) -> None:
config = AgentConfig.from_sources(
{**SECRETS, "NISSE_ROLE_EXECUTOR": "claude-haiku-4-5"}, write(tmp_path, FILE)
)
assert config.roles.executor == "claude-haiku-4-5"
def test_paths_expand_the_home_directory() -> None:
config = build()
assert config.workspace.is_absolute()
assert "~" not in str(config.state_db)
def test_an_overridden_key_needs_no_resolvable_reference(tmp_path: Path) -> None:
env = {k: v for k, v in SECRETS.items() if k != "TEST_LLM_TOKEN"}
config = AgentConfig.from_sources(
{**env, "NISSE_LLM_TOKEN": "sk-env"}, write(tmp_path, FILE)
)
assert config.llm_token == "sk-env"
def test_every_missing_setting_is_reported_at_once() -> None:
# Failing at first use would surface a missing token twenty minutes into
# a milestone, on the call that mattered.
with pytest.raises(ConfigError) as excinfo:
build(matrix__token=None, gitea__repo=None)
message = str(excinfo.value)
assert "matrix.token" in message
assert "gitea.repo" in message
def test_a_reference_to_an_unset_variable_is_an_error(tmp_path: Path) -> None:
def test_empty_is_treated_as_missing() -> None:
with pytest.raises(ConfigError, match="gitea.repo"):
build(gitea__repo=" ")
def test_a_reference_to_an_unset_variable_is_an_error() -> None:
# Never an empty string: an empty token would pass for a value and fail
# at the first request.
env = {k: v for k, v in SECRETS.items() if k != "TEST_GITEA_TOKEN"}
with pytest.raises(ConfigError, match=r"gitea.token: references \$\{TEST_GITEA"):
AgentConfig.from_sources(env, write(tmp_path, FILE))
AgentConfig.from_tree(tree(), env)
def test_tuning_is_read_from_the_file(tmp_path: Path) -> None:
text = (
FILE
+ """
tuning:
executor: {max_turns: 40, failure_limit: 3}
expert: {consult_budget: 5}
delegation: {max_turns: 20, max_tokens: 100000, wall_clock_s: 600}
tools: {command_timeout_s: 300, max_output_chars: 2000}
"""
def test_operator_lists_drop_blanks() -> None:
config = build(matrix__operators=[" @don:test ", "@other:test", " "])
assert config.operators == {"@don:test", "@other:test"}
# A blank entry must not admit an empty identity into an allowlist.
assert "" not in config.operators
def test_an_operator_list_of_only_blanks_is_missing() -> None:
with pytest.raises(ConfigError, match="gitea.operators"):
build(gitea__operators=[" "])
def test_base_branch_is_configurable() -> None:
assert build(gitea__base_branch="trunk").base_branch == "trunk"
def test_max_output_tokens_is_configurable_and_checked() -> None:
assert build(llm__max_output_tokens=8192).max_output_tokens == 8192
with pytest.raises(ConfigError, match="llm.max_output_tokens"):
build(llm__max_output_tokens=0)
def test_redacted_summary_carries_no_secrets() -> None:
summary = build().redacted()
flat = " ".join(summary.values())
for secret in SECRETS.values():
assert secret not in flat
assert summary["project"] == "greet"
def test_tuning_is_read() -> None:
config = build(
tuning={
"executor": {"max_turns": 40, "failure_limit": 3},
"expert": {"consult_budget": 5},
"delegation": {"max_turns": 20, "max_tokens": 100000, "wall_clock_s": 600},
"tools": {"command_timeout_s": 300, "max_output_chars": 2000},
}
)
tuning = AgentConfig.from_sources(SECRETS, write(tmp_path, text)).tuning
tuning = config.tuning
assert (tuning.executor.max_turns, tuning.executor.failure_limit) == (40, 3)
assert tuning.expert.consult_budget == 5
assert tuning.delegation.max_tokens == 100_000
assert tuning.delegation.wall_clock_s == 600.0
assert tuning.tools.max_output_chars == 2000
# A section that is only partly given keeps defaults for the rest.
partial = FILE + "\ntuning:\n executor: {max_turns: 40}\n"
tuning = AgentConfig.from_sources(SECRETS, write(tmp_path, partial)).tuning
def test_a_partly_given_tuning_section_keeps_the_other_defaults() -> None:
tuning = build(tuning={"executor": {"max_turns": 40}}).tuning
assert tuning.executor.failure_limit == 2
def test_every_problem_in_a_file_is_reported_at_once(tmp_path: Path) -> None:
def test_every_problem_is_reported_at_once() -> None:
# A misspelt knob that was silently ignored would look exactly like a knob
# with no effect, so unknown keys are errors.
text = (
FILE.replace("project: greet\n", "")
+ """
llm_extra: 1
tuning:
executor: {max_turn: 40, failure_limit: 0}
expert: {consult_budget: true}
bogus: {x: 1}
"""
)
bad = tree(project=None, llm_extra=1)
bad["tuning"] = {
"executor": {"max_turn": 40, "failure_limit": 0},
"expert": {"consult_budget": True},
"bogus": {"x": 1},
}
with pytest.raises(ConfigError) as excinfo:
AgentConfig.from_sources(SECRETS, write(tmp_path, text))
AgentConfig.from_tree(bad, SECRETS)
message = str(excinfo.value)
for expected in (
"project (NISSE_PROJECT)",
"project",
"llm_extra: unknown setting",
"tuning.executor.max_turn: unknown setting",
"tuning.executor.failure_limit: must be a positive number",
@@ -254,50 +175,76 @@ tuning:
assert expected in message
def test_a_whole_number_is_required_where_one_is_meant(tmp_path: Path) -> None:
text = FILE + "\ntuning:\n executor: {max_turns: 2.5}\n"
def test_a_whole_number_is_required_where_one_is_meant() -> None:
with pytest.raises(ConfigError, match="tuning.executor.max_turns"):
AgentConfig.from_sources(SECRETS, write(tmp_path, text))
build(tuning={"executor": {"max_turns": 2.5}})
def test_max_output_tokens_comes_from_the_file_or_the_environment(
tmp_path: Path,
) -> None:
text = FILE.replace(
" base_url: https://litellm.test\n",
" base_url: https://litellm.test\n max_output_tokens: 8192\n",
)
path = write(tmp_path, text)
assert AgentConfig.from_sources(SECRETS, path).max_output_tokens == 8192
env = {**SECRETS, "NISSE_MAX_OUTPUT_TOKENS": "4096"}
assert AgentConfig.from_sources(env, path).max_output_tokens == 4096
# --- the file --------------------------------------------------------------
FILE = """
state_db: ~/demo/state.db
workspace: ~/demo/work
project: greet
llm: {base_url: "https://litellm.test", token: "${TEST_LLM_TOKEN}"}
roles:
executor: local-fast
planner: claude-opus-5
delegate: claude-opus-5
consultant: claude-haiku-4-5
matrix:
homeserver: https://matrix.test
user_id: "@nisse-dev-testing:test"
token: ${TEST_MATRIX_TOKEN}
room_id: "!room:test"
operators: ["@don:test"]
gitea:
url: https://gitea.test
token: ${TEST_GITEA_TOKEN}
repo: don/demo
operators: [don]
"""
def test_load_finds_the_named_file_and_refuses_a_missing_one(tmp_path: Path) -> None:
path = write(tmp_path, FILE)
def write(directory: Path, text: str = FILE) -> Path:
path = directory / "nisse-dev.yaml"
path.write_text(text)
return path
def test_a_file_is_read(tmp_path: Path) -> None:
config = AgentConfig.from_file(write(tmp_path), SECRETS)
assert config.roles.executor == "local-fast"
assert config.config_file == tmp_path / "nisse-dev.yaml"
def test_invalid_yaml_is_reported(tmp_path: Path) -> None:
with pytest.raises(ConfigError, match="not valid YAML"):
AgentConfig.from_file(write(tmp_path, "roles: [unclosed"), SECRETS)
def test_load_uses_nisse_config(tmp_path: Path) -> None:
path = write(tmp_path)
assert AgentConfig.load({**SECRETS, "NISSE_CONFIG": str(path)}).project == "greet"
with pytest.raises(ConfigError, match="missing"):
AgentConfig.load({**SECRETS, "NISSE_CONFIG": str(tmp_path / "absent.yaml")})
def test_load_without_a_file_uses_the_environment(
def test_load_defaults_to_the_working_directory(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch
) -> None:
monkeypatch.chdir(tmp_path) # no ./nisse-dev.yaml here
assert AgentConfig.load(COMPLETE).config_file is None
write(tmp_path)
monkeypatch.chdir(tmp_path)
assert AgentConfig.load(SECRETS).project == "greet"
def test_a_config_file_inside_the_clone_is_refused(tmp_path: Path) -> None:
work = tmp_path / "project"
work.mkdir()
text = FILE.replace("workspace: ~/demo/work", f"workspace: {work}")
path = work / "nisse-dev.yaml"
path.write_text(text)
with pytest.raises(ConfigError, match="NISSE_CONFIG"):
AgentConfig.from_sources(SECRETS, path)
def test_load_without_a_file_says_how_to_make_one(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch
) -> None:
monkeypatch.chdir(tmp_path)
with pytest.raises(ConfigError, match="nisse-dev.example.yaml"):
AgentConfig.load(SECRETS)
def test_the_example_file_is_valid(tmp_path: Path) -> None:
def test_the_example_file_is_valid() -> None:
# The example documents every knob; it must not drift out of loading.
example = Path(__file__).resolve().parents[2] / "nisse-dev.example.yaml"
env = {
@@ -305,5 +252,48 @@ def test_the_example_file_is_valid(tmp_path: Path) -> None:
"NISSE_MATRIX_TOKEN": "syt_x",
"NISSE_GITEA_TOKEN": "gt",
}
config = AgentConfig.from_sources(env, example)
assert config.tuning.executor.max_turns == 24
assert AgentConfig.from_file(example, env).tuning.executor.max_turns == 24
# --- nothing that carries secrets may sit inside the clone -----------------
def test_state_db_and_workspace_are_separate() -> None:
# A2: build artefacts must not share a volume with agent state.
config = build()
assert config.state_db.parent != config.workspace
def test_state_db_inside_the_clone_is_refused(tmp_path: Path) -> None:
# git add -A would commit it, and the push would publish the agent's
# own state alongside the project.
work = tmp_path / "project"
with pytest.raises(ConfigError, match="inside the workspace"):
build(workspace=str(work), state_db=str(work / "state.db"))
def test_env_file_inside_the_clone_is_refused(tmp_path: Path) -> None:
# The severe case: the agent would commit and push its own credentials
# as a side effect of working correctly.
work = tmp_path / "project"
with pytest.raises(ConfigError, match="NISSE_ENV"):
AgentConfig.from_tree(
tree(workspace=str(work), state_db=str(tmp_path / "state.db")),
{**SECRETS, "NISSE_ENV": str(work / ".env")},
)
def test_a_config_file_inside_the_clone_is_refused(tmp_path: Path) -> None:
work = tmp_path / "project"
work.mkdir()
text = FILE.replace("workspace: ~/demo/work", f"workspace: {work}")
with pytest.raises(ConfigError, match="NISSE_CONFIG"):
AgentConfig.from_file(write(work, text), SECRETS)
def test_sibling_paths_are_fine(tmp_path: Path) -> None:
config = AgentConfig.from_tree(
tree(workspace=str(tmp_path / "project"), state_db=str(tmp_path / "state.db")),
{**SECRETS, "NISSE_ENV": str(tmp_path / ".env")},
)
assert config.workspace.name == "project"
nisse-dev/tests/test_dev_main.py
+7 -9
View File
@@ -2,25 +2,23 @@ from pathlib import Path
import pytest
from nisse_dev.config import REQUIRED
from nisse_dev.main import main
def test_main_refuses_to_start_without_configuration(
def test_main_refuses_to_start_without_a_config_file(
monkeypatch: pytest.MonkeyPatch,
capsys: pytest.CaptureFixture[str],
tmp_path: Path,
) -> None:
# Clear the real variables rather than swapping os.environ: from_env binds
# the environment object as a default argument, so a replacement would be
# ignored. And point NISSE_ENV at a file that does not exist, because the
# working directory's .env holds live credentials.
for name in REQUIRED:
monkeypatch.delenv(name, raising=False)
# Run from an empty directory, and point NISSE_ENV at a file that does not
# exist: the repository's .env holds live credentials, and a test must
# never be one cwd away from starting a real agent.
monkeypatch.chdir(tmp_path)
monkeypatch.delenv("NISSE_CONFIG", raising=False)
monkeypatch.setenv("NISSE_ENV", str(tmp_path / "absent.env"))
assert main() == 2
err = capsys.readouterr().err
assert "cannot start" in err
assert "NISSE_LLM_TOKEN" in err
assert "nisse-dev.example.yaml" in err
nisse-dev/tests/test_gitea.py
-5
View File
@@ -162,11 +162,6 @@ def test_list_comments_passes_since() -> None:
assert seen["since"] == "2026-09-07T00:00:00Z"
def test_from_env_requires_config() -> None:
with pytest.raises(GiteaError, match="NISSE_GITEA_URL"):
GiteaClient.from_env({})
# --- reconciler (A3 external truth) --------------------------------------
nisse-dev/tests/test_llm.py
+11 -23
View File
@@ -21,7 +21,6 @@ from nisse_dev.llm import (
RoleMap,
TruncatedResponse,
UpstreamUnavailable,
max_output_tokens,
)
EVENTS = [StoredEvent(id=1, ts="t", event=UserMessage(text="hi", source="m:1"))]
@@ -165,16 +164,16 @@ def test_malformed_response_raises_llm_error() -> None:
client.complete(EVENTS, tools=None, model="local/qwen")
ROLE_ENV = {
"NISSE_ROLE_EXECUTOR": "local/qwen",
"NISSE_ROLE_PLANNER": "anthropic/claude-opus-5",
"NISSE_ROLE_DELEGATE": "anthropic/claude-opus-5",
"NISSE_ROLE_CONSULTANT": "anthropic/claude-opus-5",
}
ROLES = RoleMap(
executor="local/qwen",
planner="anthropic/claude-opus-5",
delegate="anthropic/claude-opus-5",
consultant="anthropic/claude-opus-5",
)
def test_role_map_resolves_and_pins() -> None:
roles = RoleMap.from_env(ROLE_ENV)
roles = ROLES
assert roles.resolve("executor") == "local/qwen"
assert roles.pinned() == {
"executor": "local/qwen",
@@ -184,19 +183,9 @@ def test_role_map_resolves_and_pins() -> None:
}
def test_role_map_missing_env_raises() -> None:
with pytest.raises(LLMError, match="NISSE_ROLE_PLANNER"):
RoleMap.from_env({"NISSE_ROLE_EXECUTOR": "local/qwen"})
def test_unknown_role_raises() -> None:
with pytest.raises(LLMError, match="unknown role"):
RoleMap.from_env(ROLE_ENV).resolve("boss")
def test_client_from_env_requires_config() -> None:
with pytest.raises(LLMError, match="NISSE_LLM_BASE_URL"):
LiteLLMClient.from_env({})
ROLES.resolve("boss")
def test_load_env_file(tmp_path: Path) -> None:
@@ -306,12 +295,11 @@ def test_a_complete_answer_is_not_mistaken_for_truncation() -> None:
assert call.arguments["content"] == "a real plan"
def test_output_ceiling_is_configurable_and_generous_by_default() -> None:
def test_output_ceiling_is_generous_by_default() -> None:
# Output is billed by what is produced, so a low ceiling buys nothing and
# costs silent truncation.
# costs silent truncation. The configured value is tested in test_config.
assert DEFAULT_MAX_OUTPUT_TOKENS >= 8192
assert max_output_tokens({}) == DEFAULT_MAX_OUTPUT_TOKENS
assert max_output_tokens({"NISSE_MAX_OUTPUT_TOKENS": "2048"}) == 2048
assert LiteLLMClient("http://x", "t")._max_tokens == DEFAULT_MAX_OUTPUT_TOKENS
def test_truncation_is_an_llm_error_so_callers_already_handle_it() -> None:
nisse-dev/tests/test_matrix.py
-17
View File
@@ -143,20 +143,3 @@ def test_start_without_handler_raises() -> None:
)
with pytest.raises(TransportError, match="no message handler"):
transport.start()
def test_config_from_env_reports_missing() -> None:
with pytest.raises(TransportError, match="NISSE_MATRIX_TOKEN"):
MatrixConfig.from_env({"NISSE_MATRIX_HOMESERVER": "https://m"})
def test_config_from_env_custom_room_var() -> None:
env = {
"NISSE_MATRIX_HOMESERVER": "https://m/",
"NISSE_MATRIX_USER_ID": "@nisse-dev:test",
"NISSE_MATRIX_TOKEN": "t",
"NISSE_MATRIX_TEST_ROOM_ID": "!t:test",
}
config = MatrixConfig.from_env(env, room_var="NISSE_MATRIX_TEST_ROOM_ID")
assert config.room_id == "!t:test"
assert config.homeserver == "https://m" # trailing slash stripped
tests/integration/conftest.py
+62
View File
@@ -0,0 +1,62 @@
"""Shared setup for the live integration tests.
They run against real services, configured by nisse-test.yaml beside this
file: an ordinary agent config aimed at test resources, committed to the
repository. Its secrets are ${…} references, supplied by CI as environment
variables and locally by .env at the repository root.
A missing secret is a failure, not a skip. A half-configured environment must
not pass quietly as a skip: that is how a broken CI credential hides.
"""
from pathlib import Path
import pytest
from nisse_dev.config import AgentConfig
from nisse_dev.envfile import load_env_file
from nisse_dev.gitea import GiteaClient
from nisse_dev.llm import AnthropicPassthroughClient, LiteLLMClient
REPO = Path(__file__).resolve().parents[2]
TEST_CONFIG = Path(__file__).parent / "nisse-test.yaml"
# At import, so per-run switches such as NISSE_CAPABILITY_MODELS can be set
# in .env too. Real environment variables win over the file.
load_env_file(REPO / ".env")
@pytest.fixture(scope="session")
def live_config() -> AgentConfig:
return AgentConfig.from_file(TEST_CONFIG)
@pytest.fixture
def live_model(live_config: AgentConfig) -> str:
"""The model the live tests plan, execute and delegate with."""
return live_config.roles.planner
@pytest.fixture
def litellm(live_config: AgentConfig) -> LiteLLMClient:
return LiteLLMClient(
live_config.llm_base_url,
live_config.llm_token,
max_tokens=live_config.max_output_tokens,
)
@pytest.fixture
def passthrough(live_config: AgentConfig) -> AnthropicPassthroughClient:
return AnthropicPassthroughClient(
live_config.llm_base_url,
live_config.llm_token,
max_tokens=live_config.max_output_tokens,
)
@pytest.fixture
def gitea(live_config: AgentConfig) -> GiteaClient:
client = GiteaClient(live_config.gitea_url, live_config.gitea_token)
yield client
client.close()
tests/integration/nisse-test.yaml
+40
View File
@@ -0,0 +1,40 @@
# Agent config for the live integration tests, aimed at test resources: the
# testing identities, the dedicated test room, and the sandbox repo. Never
# point it at production.
#
# Secrets are ${…} references. CI passes them as environment variables;
# locally they come from .env at the repository root. A missing one fails
# the live tests that need it, rather than skipping them, so a broken
# credential cannot hide as a skip.
# Unused by the live tests, which work in pytest's temporary directories;
# present because every agent config has them.
state_db: /tmp/nisse-test/state.db
workspace: /tmp/nisse-test/work
project: nisse-dev-testing
llm:
base_url: https://litellm.dongar.net
token: ${NISSE_LLM_TOKEN}
# The live tests exercise the gated loop and the routes over a real proxy,
# not local-model competence (measured separately by test_model_capability),
# so every role uses a model that holds the tool-call contract.
roles:
executor: claude-haiku-4-5
planner: claude-haiku-4-5
delegate: claude-haiku-4-5
consultant: claude-haiku-4-5
matrix:
homeserver: https://matrix.dongar.net
user_id: "@nisse-dev-testing:dongar.net"
token: ${NISSE_MATRIX_TOKEN}
room_id: "!UxpXhTlQsFIeNZwkgv:dongar.net" # the test room, never production
operators: ["@dongar:dongar.net"]
gitea:
url: https://gitea.dongar.net
token: ${NISSE_GITEA_TOKEN}
repo: dongar/nisse-dev-testing-sandbox # the sandbox, never a real project
operators: [dongar]
tests/integration/test_delegation_live.py
+18 -20
View File
@@ -2,30 +2,18 @@
working code and a structured handback out."""
import io
import os
from pathlib import Path
import pytest
from nisse_dev.delegation import Caps, Delegation
from nisse_dev.envfile import load_env_file
from nisse_dev.git import GitRepo
from nisse_dev.llm import AnthropicPassthroughClient
from nisse_dev.oplog import OpLog
from nisse_dev.state import StateStore
from nisse_dev.tools import Workspace
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_LIVE_PLANNER_MODEL"
pytestmark = [
pytest.mark.integration,
pytest.mark.skipif(
not os.environ.get(SENTINEL),
reason=f"no live frontier route configured ({SENTINEL} unset)",
),
]
pytestmark = pytest.mark.integration
BRIEF = """Create `slugify.py` in the workspace root containing a function
`slugify(text)` that lowercases the text and replaces every run of
@@ -34,7 +22,9 @@ hyphen. Then create `test_slugify.py` with at least two assertions covering
it, and run `python -m pytest test_slugify.py -q` until it passes."""
def build(tmp_path: Path, caps: Caps) -> tuple[Delegation, GitRepo, StateStore]:
def build(
tmp_path: Path, caps: Caps, client: AnthropicPassthroughClient, model: str
) -> tuple[Delegation, GitRepo, StateStore]:
repo = GitRepo(tmp_path / "work")
repo.init()
repo.configure_identity("nisse-dev-testing", "n@t.test")
@@ -42,18 +32,22 @@ def build(tmp_path: Path, caps: Caps) -> tuple[Delegation, GitRepo, StateStore]:
repo.commit_all("seed")
store = StateStore(tmp_path / "state.db", oplog=OpLog("nisse-dev", io.StringIO()))
delegation = Delegation(
AnthropicPassthroughClient.from_env(),
client,
store,
Workspace(repo.root, timeout_s=60),
repo,
model=os.environ[SENTINEL],
model=model,
caps=caps,
)
return delegation, repo, store
def test_delegate_completes_a_scoped_subtask(tmp_path: Path) -> None:
delegation, repo, _store = build(tmp_path, Caps(turns=12, wall_clock_s=300.0))
def test_delegate_completes_a_scoped_subtask(
tmp_path: Path, passthrough: AnthropicPassthroughClient, live_model: str
) -> None:
delegation, repo, _store = build(
tmp_path, Caps(turns=12, wall_clock_s=300.0), passthrough, live_model
)
result = delegation.run(brief=BRIEF, branch="live/delegation-1")
@@ -76,10 +70,14 @@ def test_delegate_completes_a_scoped_subtask(tmp_path: Path) -> None:
assert check.success, f"delegate's own tests do not pass: {check.result[:400]}"
def test_a_cap_breach_discards_the_work(tmp_path: Path) -> None:
def test_a_cap_breach_discards_the_work(
tmp_path: Path, passthrough: AnthropicPassthroughClient, live_model: str
) -> None:
# One turn cannot finish the brief, so this exercises the abort path
# against a real model rather than a scripted one.
delegation, repo, _ = build(tmp_path, Caps(turns=1, wall_clock_s=120.0))
delegation, repo, _ = build(
tmp_path, Caps(turns=1, wall_clock_s=120.0), passthrough, live_model
)
result = delegation.run(brief=BRIEF, branch="live/delegation-2")
tests/integration/test_entry_points.py
+11 -6
View File
@@ -2,6 +2,7 @@
import os
import subprocess
from pathlib import Path
import pytest
@@ -16,27 +17,31 @@ def test_ops_entry_point_runs_and_greets() -> None:
assert result.stdout == "Hello from nisse-ops, the house spirit is awake.\n"
def test_dev_entry_point_refuses_to_start_unconfigured() -> None:
"""Missing configuration must fail at startup, naming everything absent.
def test_dev_entry_point_refuses_to_start_unconfigured(tmp_path: Path) -> None:
"""An incomplete config must fail at startup, naming everything absent.
Run with a deliberately empty environment and a .env path that does not
exist: the real .env in the working directory holds live credentials, and
a test must never be one cwd away from starting a real agent.
Run from an empty directory, with an empty config file and a .env path
that does not exist: the repository's .env holds live credentials, and a
test must never be one cwd away from starting a real agent.
"""
config = tmp_path / "nisse-dev.yaml"
config.write_text("")
result = subprocess.run(
["nisse-dev"],
capture_output=True,
text=True,
timeout=30,
check=False,
cwd=tmp_path,
env={
"PATH": os.environ["PATH"],
"HOME": os.environ["HOME"],
"NISSE_ENV": "/nonexistent/.env",
"NISSE_CONFIG": str(config),
},
)
assert result.returncode == 2
assert "cannot start" in result.stderr
# All of them at once, not the first one it happened to reach.
for expected in ("NISSE_STATE_DB", "NISSE_MATRIX_TOKEN", "NISSE_GITEA_REPO"):
for expected in ("state_db", "matrix.token", "gitea.repo"):
assert expected in result.stderr
tests/integration/test_expert_live.py
+16 -22
View File
@@ -5,40 +5,34 @@ that parses against our local schema, since no API-level enforcement exists.
"""
import io
import os
from pathlib import Path
import pytest
from nisse_dev.envfile import load_env_file
from nisse_dev.evals import collect
from nisse_dev.expert import BudgetExhausted, ExpertDesk
from nisse_dev.llm import LiteLLMClient, RoleMap
from nisse_dev.oplog import OpLog
from nisse_dev.state import StateStore
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_LIVE_PLANNER_MODEL"
pytestmark = [
pytest.mark.integration,
pytest.mark.skipif(
not os.environ.get(SENTINEL),
reason=f"no live frontier route configured ({SENTINEL} unset)",
),
]
pytestmark = pytest.mark.integration
def desk(tmp_path: Path, budget: int = 2) -> tuple[ExpertDesk, StateStore]:
model = os.environ[SENTINEL]
roles = RoleMap(executor=model, planner=model, delegate=model, consultant=model)
store = StateStore(tmp_path / "state.db", oplog=OpLog("nisse-dev", io.StringIO()))
return ExpertDesk(LiteLLMClient.from_env(), store, roles, budget=budget), store
@pytest.fixture
def desk(tmp_path: Path, litellm: LiteLLMClient, live_model: str):
def make(budget: int = 2) -> tuple[ExpertDesk, StateStore]:
model = live_model
roles = RoleMap(executor=model, planner=model, delegate=model, consultant=model)
store = StateStore(
tmp_path / "state.db", oplog=OpLog("nisse-dev", io.StringIO())
)
return ExpertDesk(litellm, store, roles, budget=budget), store
return make
def test_live_consult_parses_against_the_local_schema(tmp_path: Path) -> None:
expert, store = desk(tmp_path)
def test_live_consult_parses_against_the_local_schema(desk) -> None:
expert, store = desk()
answer = expert.ask(
"For a single-process agent storing a few thousand rows of task state, "
@@ -53,8 +47,8 @@ def test_live_consult_parses_against_the_local_schema(tmp_path: Path) -> None:
assert record.exit_status == "answered"
def test_live_budget_stops_a_circling_caller(tmp_path: Path) -> None:
expert, _ = desk(tmp_path, budget=1)
def test_live_budget_stops_a_circling_caller(desk) -> None:
expert, _ = desk(budget=1)
expert.ask("Name one advantage of WAL mode in SQLite. Be brief.")
with pytest.raises(BudgetExhausted):
expert.ask("And another?")
tests/integration/test_gitea_live.py
+9 -24
View File
@@ -5,29 +5,17 @@ Cleans up after itself (closes the PR, deletes the branch) so repeated runs
do not silt up the sandbox.
"""
import os
import uuid
from pathlib import Path
import pytest
from nisse_dev.envfile import load_env_file
from nisse_dev.config import AgentConfig
from nisse_dev.git import GitError, GitRepo, authenticated_url
from nisse_dev.gitea import GiteaClient, pull_request_reconciler
from nisse_dev.state import Intent
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_GITEA_URL"
ALSO_REQUIRED = ("NISSE_GITEA_TOKEN", "NISSE_GITEA_TEST_REPO")
pytestmark = [
pytest.mark.integration,
pytest.mark.skipif(
not os.environ.get(SENTINEL),
reason=f"no live Gitea configured ({SENTINEL} unset)",
),
]
pytestmark = pytest.mark.integration
def intent_for(branch: str) -> Intent:
@@ -41,16 +29,13 @@ def intent_for(branch: str) -> Intent:
)
def test_push_open_pr_and_reconcile(tmp_path: Path) -> None:
missing = [var for var in ALSO_REQUIRED if not os.environ.get(var)]
assert not missing, (
f"{SENTINEL} is set, so the live Gitea path is expected to work, "
f"but these are missing: {', '.join(missing)}"
)
base_url = os.environ[SENTINEL]
token = os.environ["NISSE_GITEA_TOKEN"]
repo_path = os.environ["NISSE_GITEA_TEST_REPO"]
client = GiteaClient.from_env()
def test_push_open_pr_and_reconcile(
tmp_path: Path, live_config: AgentConfig, gitea: GiteaClient
) -> None:
base_url = live_config.gitea_url
token = live_config.gitea_token
repo_path = live_config.gitea_repo # the sandbox
client = gitea
account = client.whoami()
assert "testing" in account, (
tests/integration/test_llm_live.py
+7 -33
View File
@@ -1,41 +1,17 @@
"""Live smoke test against the real LiteLLM proxy.
NISSE_LLM_BASE_URL alone decides whether this runs: if it is set, the live
path is expected to work, and anything else missing is a *failure*, not a
skip. A half-configured environment must not pass quietly as a skip — that is
how a broken CI credential hides.
"""
import os
from pathlib import Path
"""Live smoke test against the real LiteLLM proxy."""
import pytest
from nisse_dev.canonical import AssistantText, StoredEvent, UserMessage
from nisse_dev.envfile import load_env_file
from nisse_dev.config import AgentConfig
from nisse_dev.llm import LiteLLMClient
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_LLM_BASE_URL"
ALSO_REQUIRED = ("NISSE_LLM_TOKEN", "NISSE_ROLE_EXECUTOR")
pytestmark = [
pytest.mark.integration,
pytest.mark.skipif(
not os.environ.get(SENTINEL),
reason=f"no live LLM configured ({SENTINEL} unset)",
),
]
pytestmark = pytest.mark.integration
def test_executor_model_completes_via_litellm() -> None:
missing = [var for var in ALSO_REQUIRED if not os.environ.get(var)]
assert not missing, (
f"{SENTINEL} is set, so the live LLM path is expected to work, "
f"but these are missing: {', '.join(missing)}"
)
client = LiteLLMClient.from_env()
def test_executor_model_completes_via_litellm(
litellm: LiteLLMClient, live_config: AgentConfig
) -> None:
events = [
StoredEvent(
id=1,
@@ -45,9 +21,7 @@ def test_executor_model_completes_via_litellm() -> None:
),
)
]
result = client.complete(
events, tools=None, model=os.environ["NISSE_ROLE_EXECUTOR"]
)
result = litellm.complete(events, tools=None, model=live_config.roles.executor)
texts = [e.text for e in result if isinstance(e, AssistantText)]
assert texts, f"no assistant text in {result!r}"
assert "pong" in texts[0].lower()
tests/integration/test_matrix_live.py
+6 -33
View File
@@ -1,36 +1,14 @@
"""Live smoke test against the real homeserver, bound strictly to the
dedicated test room (never the production room).
dedicated test room that nisse-test.yaml names (never the production room)."""
NISSE_MATRIX_HOMESERVER alone decides whether this runs: if it is set, the
live path is expected to work, and anything else missing is a *failure*, not
a skip. A half-configured environment must not pass quietly as a skip — that
is how a broken CI credential hides."""
import os
import uuid
from pathlib import Path
import pytest
from nisse_dev.envfile import load_env_file
from nisse_dev.config import AgentConfig
from nisse_dev.matrix import MatrixConfig, MatrixTransport
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_MATRIX_HOMESERVER"
ALSO_REQUIRED = (
"NISSE_MATRIX_USER_ID",
"NISSE_MATRIX_TOKEN",
"NISSE_MATRIX_TEST_ROOM_ID",
)
pytestmark = [
pytest.mark.integration,
pytest.mark.skipif(
not os.environ.get(SENTINEL),
reason=f"no live homeserver configured ({SENTINEL} unset)",
),
]
pytestmark = pytest.mark.integration
def _assert_testing_identity(transport: MatrixTransport, config: MatrixConfig) -> None:
@@ -46,7 +24,7 @@ def _assert_testing_identity(transport: MatrixTransport, config: MatrixConfig) -
whoami.raise_for_status()
actual = whoami.json()["user_id"]
assert actual == config.user_id, (
f"NISSE_MATRIX_USER_ID is {config.user_id} but the token belongs to "
f"matrix.user_id is {config.user_id} but the token belongs to "
f"{actual}; the self-message filter would not match"
)
assert "testing" in actual, (
@@ -54,13 +32,8 @@ def _assert_testing_identity(transport: MatrixTransport, config: MatrixConfig) -
)
def test_send_lands_in_test_room() -> None:
missing = [var for var in ALSO_REQUIRED if not os.environ.get(var)]
assert not missing, (
f"{SENTINEL} is set, so the live Matrix path is expected to work, "
f"but these are missing: {', '.join(missing)}"
)
config = MatrixConfig.from_env(room_var="NISSE_MATRIX_TEST_ROOM_ID")
def test_send_lands_in_test_room(live_config: AgentConfig) -> None:
config = live_config.matrix
marker = f"transport smoke test {uuid.uuid4().hex}"
transport = MatrixTransport(config)
try:
tests/integration/test_milestone_live.py
+8 -22
View File
@@ -2,12 +2,11 @@
approve through the gate, then let the executor role actually create and test
a file on disk.
Gated on NISSE_LIVE_PLANNER_MODEL like the planning test — it needs a model
that holds the tool-call contract on both ends.
Uses the test config's planner model for every role: it needs a model that
holds the tool-call contract on both ends.
"""
import io
import os
from pathlib import Path
import pytest
@@ -19,7 +18,6 @@ from nisse_dev.control import (
EXECUTING,
Orchestrator,
)
from nisse_dev.envfile import load_env_file
from nisse_dev.executor import LLMExecutor
from nisse_dev.llm import LiteLLMClient, RoleMap
from nisse_dev.oplog import OpLog
@@ -28,17 +26,7 @@ from nisse_dev.state import StateStore
from nisse_dev.tools import Workspace
from nisse_dev.transport import InMemoryTransport
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_LIVE_PLANNER_MODEL"
pytestmark = [
pytest.mark.integration,
pytest.mark.skipif(
not os.environ.get(SENTINEL),
reason=f"no live planner route configured ({SENTINEL} unset)",
),
]
pytestmark = pytest.mark.integration
def _drive_to_execution(
@@ -67,16 +55,14 @@ def _drive_to_execution(
)
def test_plan_approve_execute_produces_a_real_file(tmp_path: Path) -> None:
assert os.environ.get("NISSE_LLM_TOKEN"), (
f"{SENTINEL} is set, so the live path is expected to work, "
"but NISSE_LLM_TOKEN is missing"
)
def test_plan_approve_execute_produces_a_real_file(
tmp_path: Path, litellm: LiteLLMClient, live_model: str
) -> None:
# Both roles use the capable model here: the point is the whole gated
# loop over a real proxy, not local-model competence (measured separately).
model = os.environ[SENTINEL]
model = live_model
roles = RoleMap(executor=model, planner=model, delegate=model, consultant=model)
client = LiteLLMClient.from_env()
client = litellm
store = StateStore(tmp_path / "state.db", oplog=OpLog("nisse-dev", io.StringIO()))
workspace = Workspace(tmp_path / "work", timeout_s=60)
transport = InMemoryTransport()
tests/integration/test_model_capability.py
+24 -20
View File
@@ -14,7 +14,6 @@ because it spends real tokens across several models.
"""
import os
from pathlib import Path
import pytest
@@ -26,14 +25,12 @@ from nisse_dev.canonical import (
UserMessage,
)
from nisse_dev.catalog import ModelCatalog, select_client
from nisse_dev.envfile import load_env_file
from nisse_dev.config import AgentConfig
from nisse_dev.expert import parse_answer
from nisse_dev.llm import AnthropicPassthroughClient, LiteLLMClient, LLMError
from nisse_dev.planner import PLANNER_TOOLS
from nisse_dev.tools import EXECUTOR_TOOLS
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_CAPABILITY_MODELS"
MODELS = [m.strip() for m in os.environ.get(SENTINEL, "").split(",") if m.strip()]
@@ -45,17 +42,18 @@ pytestmark = [
]
def clients() -> tuple[LiteLLMClient, AnthropicPassthroughClient]:
return LiteLLMClient.from_env(), AnthropicPassthroughClient.from_env()
@pytest.fixture
def routes(
litellm: LiteLLMClient, passthrough: AnthropicPassthroughClient
) -> tuple[LiteLLMClient, AnthropicPassthroughClient]:
return litellm, passthrough
@pytest.fixture(scope="module")
def catalog() -> ModelCatalog:
def catalog(live_config: AgentConfig) -> ModelCatalog:
"""The live catalog this key can see. /model/info is scoped per key, so
the probe list and the key have to agree."""
return ModelCatalog.discover(
os.environ["NISSE_LLM_BASE_URL"], os.environ["NISSE_LLM_TOKEN"]
)
return ModelCatalog.discover(live_config.llm_base_url, live_config.llm_token)
def user(text: str) -> list[StoredEvent]:
@@ -63,8 +61,8 @@ def user(text: str) -> list[StoredEvent]:
@pytest.mark.parametrize("model", MODELS)
def test_model_answers_at_all(model: str, catalog: ModelCatalog) -> None:
normalized, passthrough = clients()
def test_model_answers_at_all(model: str, catalog: ModelCatalog, routes) -> None:
normalized, passthrough = routes
client = select_client(model, normalized, passthrough, catalog)
produced = client.complete(
user("Reply with exactly one word: pong"), None, model, system="Be terse."
@@ -74,10 +72,12 @@ def test_model_answers_at_all(model: str, catalog: ModelCatalog) -> None:
@pytest.mark.parametrize("model", MODELS)
def test_model_holds_the_planner_contract(model: str, catalog: ModelCatalog) -> None:
def test_model_holds_the_planner_contract(
model: str, catalog: ModelCatalog, routes
) -> None:
"""A4's gating is structural at the tool layer: a model that answers in
prose cannot be gated, because there is no plan to approve."""
normalized, passthrough = clients()
normalized, passthrough = routes
client = select_client(model, normalized, passthrough, catalog)
produced = client.complete(
user("Draft an architecture plan for a command-line slug generator."),
@@ -101,10 +101,12 @@ def test_model_holds_the_planner_contract(model: str, catalog: ModelCatalog) ->
@pytest.mark.parametrize("model", MODELS)
def test_model_round_trips_a_tool_result(model: str, catalog: ModelCatalog) -> None:
def test_model_round_trips_a_tool_result(
model: str, catalog: ModelCatalog, routes
) -> None:
"""The executor loop feeds results back; a model that cannot consume its
own tool result cannot run a milestone."""
normalized, passthrough = clients()
normalized, passthrough = routes
client = select_client(model, normalized, passthrough, catalog)
events = user("List the files in the current directory using your tools.")
produced = client.complete(events, EXECUTOR_TOOLS, model, system="Use a tool.")
@@ -130,11 +132,11 @@ def test_model_round_trips_a_tool_result(model: str, catalog: ModelCatalog) -> N
@pytest.mark.parametrize("model", MODELS)
def test_model_holds_the_prompt_level_json_contract(
model: str, catalog: ModelCatalog
model: str, catalog: ModelCatalog, routes
) -> None:
"""A14 contracts structured output at the prompt level because vendors
differ on JSON enforcement. This measures whether that holds per model."""
normalized, passthrough = clients()
normalized, passthrough = routes
client = select_client(model, normalized, passthrough, catalog)
produced = client.complete(
user("Is SQLite sufficient for a single-process agent's task state?"),
@@ -153,11 +155,13 @@ def test_model_holds_the_prompt_level_json_contract(
@pytest.mark.parametrize("model", MODELS)
def test_route_selection_matches_the_model(model: str, catalog: ModelCatalog) -> None:
def test_route_selection_matches_the_model(
model: str, catalog: ModelCatalog, routes
) -> None:
"""A non-Anthropic model must not be sent Anthropic-shaped bodies. This
was hardcoded until gpt-6-astra made it observable, and name-sniffed until
semantic aliases made the name stop meaning anything."""
normalized, passthrough = clients()
normalized, passthrough = routes
facts = catalog.facts(model)
chosen = select_client(model, normalized, passthrough, catalog)
if facts.provider == "anthropic":
tests/integration/test_passthrough_live.py
+19 -26
View File
@@ -6,30 +6,19 @@ passthrough rather than being flattened by the proxy.
"""
import io
import os
from pathlib import Path
import httpx
import pytest
from nisse_dev.canonical import AssistantText, ToolCall, ToolResult, UserMessage
from nisse_dev.envfile import load_env_file
from nisse_dev.config import AgentConfig
from nisse_dev.llm import AnthropicPassthroughClient
from nisse_dev.oplog import OpLog
from nisse_dev.state import StateStore
from nisse_dev.tools import EXECUTOR_TOOLS
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_LIVE_PLANNER_MODEL"
pytestmark = [
pytest.mark.integration,
pytest.mark.skipif(
not os.environ.get(SENTINEL),
reason=f"no live frontier route configured ({SENTINEL} unset)",
),
]
pytestmark = pytest.mark.integration
def store_with(tmp_path: Path, *events) -> StateStore:
@@ -39,18 +28,22 @@ def store_with(tmp_path: Path, *events) -> StateStore:
return store
def test_passthrough_completes_a_plain_turn(tmp_path: Path) -> None:
def test_passthrough_completes_a_plain_turn(
tmp_path: Path, passthrough: AnthropicPassthroughClient, live_model: str
) -> None:
store = store_with(
tmp_path, UserMessage(text="Reply with exactly: pong", source="live")
)
result = AnthropicPassthroughClient.from_env().complete(
store.read_events(), None, os.environ[SENTINEL], system="Be terse."
result = passthrough.complete(
store.read_events(), None, live_model, system="Be terse."
)
texts = [e.text for e in result if isinstance(e, AssistantText)]
assert texts and "pong" in texts[0].lower()
def test_passthrough_round_trips_a_tool_call(tmp_path: Path) -> None:
def test_passthrough_round_trips_a_tool_call(
tmp_path: Path, passthrough: AnthropicPassthroughClient, live_model: str
) -> None:
store = store_with(
tmp_path,
UserMessage(
@@ -58,11 +51,11 @@ def test_passthrough_round_trips_a_tool_call(tmp_path: Path) -> None:
source="live",
),
)
client = AnthropicPassthroughClient.from_env()
client = passthrough
produced = client.complete(
store.read_events(),
EXECUTOR_TOOLS,
os.environ[SENTINEL],
live_model,
system="Use a tool. Do not answer in prose.",
)
calls = [e for e in produced if isinstance(e, ToolCall)]
@@ -75,18 +68,18 @@ def test_passthrough_round_trips_a_tool_call(tmp_path: Path) -> None:
store.append_event(
ToolResult(call_id=calls[0].call_id, result="app.py\ntests/", success=True)
)
follow_up = client.complete(
store.read_events(), EXECUTOR_TOOLS, os.environ[SENTINEL]
)
follow_up = client.complete(store.read_events(), EXECUTOR_TOOLS, live_model)
assert follow_up, "the vendor rejected or ignored the tool_result turn"
def test_prompt_caching_survives_the_passthrough() -> None:
def test_prompt_caching_survives_the_passthrough(
live_config: AgentConfig, live_model: str
) -> None:
# A5 asks for this to be verified rather than assumed: a proxy that
# flattened the request would silently drop cache control and turn a
# cost optimisation into a per-call expense.
base = os.environ["NISSE_LLM_BASE_URL"].rstrip("/")
token = os.environ["NISSE_LLM_TOKEN"]
base = live_config.llm_base_url.rstrip("/")
token = live_config.llm_token
response = httpx.post(
f"{base}/v1/messages",
headers={
@@ -95,7 +88,7 @@ def test_prompt_caching_survives_the_passthrough() -> None:
"anthropic-version": "2023-06-01",
},
json={
"model": os.environ[SENTINEL],
"model": live_model,
"max_tokens": 16,
"messages": [{"role": "user", "content": "Say pong."}],
},
tests/integration/test_planning_live.py
+9 -26
View File
@@ -1,47 +1,30 @@
"""Live planning turn: a real model must answer with a submit_plan or
ask_human tool call, not prose.
Deliberately opt-in via NISSE_LIVE_PLANNER_MODEL rather than the usual
NISSE_LLM_BASE_URL sentinel. Planning escalates to a frontier model
unconditionally (A12), and the small local model does not hold the tool-call
contract reliably — measured, see the PR — so this test needs a frontier
route that does not exist yet. Set the variable to the model to plan with
once one is configured; until then this skips on purpose, not by accident.
Plans with the test config's planner role. Planning escalates to a frontier
model unconditionally (A12), and small local models do not hold the tool-call
contract reliably, so that role must be one that does.
"""
import io
import os
from pathlib import Path
import pytest
from nisse_dev.canonical import UserMessage
from nisse_dev.control import AskHuman, SubmitPlan
from nisse_dev.envfile import load_env_file
from nisse_dev.llm import LiteLLMClient, RoleMap
from nisse_dev.oplog import OpLog
from nisse_dev.planner import LLMPlanningAgent
from nisse_dev.state import StateStore
load_env_file(Path(__file__).resolve().parents[2] / ".env")
SENTINEL = "NISSE_LIVE_PLANNER_MODEL"
pytestmark = [
pytest.mark.integration,
pytest.mark.skipif(
not os.environ.get(SENTINEL),
reason=f"no frontier planner route configured ({SENTINEL} unset)",
),
]
pytestmark = pytest.mark.integration
def test_live_planning_turn_returns_a_tool_action(tmp_path: Path) -> None:
assert os.environ.get("NISSE_LLM_TOKEN"), (
f"{SENTINEL} is set, so the live planning path is expected to work, "
"but NISSE_LLM_TOKEN is missing"
)
model = os.environ[SENTINEL]
def test_live_planning_turn_returns_a_tool_action(
tmp_path: Path, litellm: LiteLLMClient, live_model: str
) -> None:
model = live_model
roles = RoleMap(executor=model, planner=model, delegate=model, consultant=model)
store = StateStore(tmp_path / "state.db", oplog=OpLog("nisse-dev", io.StringIO()))
store.append_event(
@@ -53,7 +36,7 @@ def test_live_planning_turn_returns_a_tool_action(tmp_path: Path) -> None:
source="live-test",
)
)
agent = LLMPlanningAgent(LiteLLMClient.from_env(), store, roles)
agent = LLMPlanningAgent(litellm, store, roles)
action = agent.plan("arch", "hello-cli")