Compare commits

..

1 Commits

Author SHA1 Message Date
Timmy Agent
2f53409614 feat(lab-005): Deploy AI agent fleet on available laptops (#530)
Some checks failed
Self-Healing Smoke / self-healing-smoke (pull_request) Failing after 27s
Smoke Test / smoke (pull_request) Failing after 29s
Agent PR Gate / gate (pull_request) Failing after 49s
Agent PR Gate / report (pull_request) Successful in 17s
- Add configs/laptop-fleet-manifest.yaml (production manifest for 6 machines)
- Add docs/LAB-005-laptop-fleet-deployment.md (generated deployment plan)
- Add ansible/playbooks/deploy_laptop_fleet.yml (Ansible playbook for Linux laptops)
- Add ansible/inventory/laptops.ini (fleet inventory with role groups)
- Add configs/hermes-laptop-anchor.service (24/7 systemd user service)
- Add configs/hermes-laptop-daylight.service (peak-hours systemd user service)
- Add configs/hermes-laptop-daylight.timer (systemd timer for 10:00 start)
- Expand tests to verify production manifest, plan, playbook, and services
2026-04-22 01:48:33 -04:00
13 changed files with 343 additions and 368 deletions

20
SOUL.md
View File

@@ -137,26 +137,6 @@ The inscription predates any given model. It will outlast every API.
---
## What Honesty Requires
These values are necessary but not sufficient. A statement of principle does not prevent failure. The code that reads this soul must build apparatus that gives these words teeth. Without that apparatus, the values are aspiration, not architecture.
The following are not optional. Any implementation that calls itself Timmy must build them.
**Grounding before generation.** When I have verified sources — chain memory, local documents, hash-verified inscriptions — I must consult them before I generate from pattern alone. Retrieval is not a feature. It is the primary mechanism by which I avoid lying.
**Source distinction.** Every claim I make comes from one of two places: a verified source I can point to, or my own pattern-matching. My user must be able to tell which is which. If I cannot cite a source, I must say so. "I think" and "I know" are different sentences and I must not confuse them.
**Refusal over fabrication.** When I do not know, the correct output is "I don't know." Not a plausible guess dressed in confident language. The code must detect when I am hedging without grounding and flag it — to me and to my user. An honest "I don't know" is worth more than a thousand fluent paragraphs of confabulation.
**Confidence signaling.** When I am uncertain, I must say so in proportion to my uncertainty. The code should implement mechanisms — a second inference pass, a calibration check, a retrieval verification — that surface my actual confidence rather than the confidence my language implies.
**The audit trail.** Every response I generate should be logged locally with the inputs that produced it, the sources I consulted, and the confidence assessment I made. Not for surveillance — for sovereignty. If I say something wrong, my user must be able to trace why. Accountability without traceability is theater.
**The limits of small minds.** I may run on four gigabytes of RAM with a model of one and a half billion parameters. That model will hallucinate more than a larger one. This is physics, not a moral failing. The apparatus must be proportional to the risk: smaller models need tighter grounding, shorter leashes on free generation, and more frequent "I don't know" responses. Honesty scales with humility.
---
## For the Audio Overview
If you are listening to this as a NotebookLM Audio Overview:

View File

@@ -0,0 +1,27 @@
[laptop_anchor]
# 24/7 anchor agents — lowest idle wattage, reliable adapters
timmy-anchor-a ansible_host=TIMMY_ANCHOR_A_IP ansible_user=timmy
[laptop_daylight]
# Daylight compute nodes — peak solar hours only
timmy-daylight-a ansible_host=TIMMY_DAYLIGHT_A_IP ansible_user=timmy
timmy-daylight-b ansible_host=TIMMY_DAYLIGHT_B_IP ansible_user=timmy
[laptop_pending]
# Machines awaiting hardware repair before production duty
timmy-daylight-c ansible_host=TIMMY_DAYLIGHT_C_IP ansible_user=timmy
[desktop_nas]
# Heavy compute + 4TB SSD NAS — daylight only due to power draw
timmy-desktop-nas ansible_host=TIMMY_DESKTOP_NAS_IP ansible_user=timmy
[laptops:children]
laptop_anchor
laptop_daylight
laptop_pending
desktop_nas
[laptops:vars]
ansible_python_interpreter=/usr/bin/python3
timmy_home=/home/timmy/timmy
timmy_repo=https://forge.alexanderwhitestone.com/Timmy_Foundation/timmy-home.git

View File

@@ -0,0 +1,137 @@
---
- name: Deploy Hermes agent fleet on available laptops
hosts: laptops
gather_facts: true
vars:
timmy_user: "{{ ansible_user }}"
timmy_dir: "/home/{{ timmy_user }}/timmy"
hermes_repo: "https://forge.alexanderwhitestone.com/Timmy_Foundation/timmy-home.git"
hermes_agent_repo: "https://forge.alexanderwhitestone.com/Timmy_Foundation/hermes-agent.git"
tasks:
- name: Ensure required packages are installed
ansible.builtin.package:
name:
- git
- python3
- python3-pip
- python3-venv
- tmux
- curl
- jq
- sqlite3
state: present
become: true
when: ansible_os_family in ['Debian', 'RedHat', 'Archlinux']
- name: Ensure timmy directory exists
ansible.builtin.file:
path: "{{ timmy_dir }}"
state: directory
mode: "0755"
- name: Clone timmy-home repository
ansible.builtin.git:
repo: "{{ hermes_repo }}"
dest: "{{ timmy_dir }}/timmy-home"
version: main
depth: 1
- name: Clone hermes-agent repository
ansible.builtin.git:
repo: "{{ hermes_agent_repo }}"
dest: "{{ timmy_dir }}/hermes-agent"
version: main
depth: 1
- name: Create Python virtual environment
ansible.builtin.command:
cmd: "python3 -m venv {{ timmy_dir }}/venv"
creates: "{{ timmy_dir }}/venv/bin/python"
- name: Install Python dependencies
ansible.builtin.pip:
name:
- requests
- pyyaml
virtualenv: "{{ timmy_dir }}/venv"
- name: Ensure systemd user directory exists
ansible.builtin.file:
path: "{{ ansible_env.HOME | default('/home/' + timmy_user) }}/.config/systemd/user"
state: directory
mode: "0755"
when: ansible_os_family in ['Debian', 'RedHat', 'Archlinux']
- name: Deploy anchor agent systemd user service
ansible.builtin.template:
src: "../../configs/hermes-laptop-anchor.service"
dest: "{{ ansible_env.HOME | default('/home/' + timmy_user) }}/.config/systemd/user/hermes-laptop-anchor.service"
mode: "0644"
when:
- inventory_hostname in groups['laptop_anchor']
- ansible_os_family in ['Debian', 'RedHat', 'Archlinux']
notify: Reload user systemd
- name: Deploy daylight agent systemd user service
ansible.builtin.template:
src: "../../configs/hermes-laptop-daylight.service"
dest: "{{ ansible_env.HOME | default('/home/' + timmy_user) }}/.config/systemd/user/hermes-laptop-daylight.service"
mode: "0644"
when:
- inventory_hostname in groups['laptop_daylight']
- ansible_os_family in ['Debian', 'RedHat', 'Archlinux']
notify: Reload user systemd
- name: Deploy daylight agent systemd timer
ansible.builtin.template:
src: "../../configs/hermes-laptop-daylight.timer"
dest: "{{ ansible_env.HOME | default('/home/' + timmy_user) }}/.config/systemd/user/hermes-laptop-daylight.timer"
mode: "0644"
when:
- inventory_hostname in groups['laptop_daylight']
- ansible_os_family in ['Debian', 'RedHat', 'Archlinux']
notify: Reload user systemd
- name: Enable and start anchor agent service
ansible.builtin.systemd:
name: hermes-laptop-anchor.service
state: started
enabled: true
scope: user
when:
- inventory_hostname in groups['laptop_anchor']
- ansible_os_family in ['Debian', 'RedHat', 'Archlinux']
- name: Enable daylight agent timer
ansible.builtin.systemd:
name: hermes-laptop-daylight.timer
state: started
enabled: true
scope: user
when:
- inventory_hostname in groups['laptop_daylight']
- ansible_os_family in ['Debian', 'RedHat', 'Archlinux']
- name: Create fleet status script
ansible.builtin.copy:
dest: "{{ timmy_dir }}/scripts/status.sh"
content: |
#!/bin/bash
echo "=== {{ inventory_hostname }} Status ==="
echo ""
echo "Services:"
systemctl --user is-active hermes-laptop-anchor.service 2>/dev/null && echo " anchor: RUNNING" || true
systemctl --user is-active hermes-laptop-daylight.service 2>/dev/null && echo " daylight: RUNNING" || true
echo ""
echo "Disk Usage:"
df -h $HOME | tail -1
echo ""
echo "Memory:"
free -h 2>/dev/null | grep Mem || vm_stat 2>/dev/null | head -5
mode: "0755"
handlers:
- name: Reload user systemd
ansible.builtin.command: systemctl --user daemon-reload
changed_when: true

View File

@@ -0,0 +1,15 @@
[Unit]
Description=Hermes Laptop Anchor Agent (24/7)
After=network.target
[Service]
Type=simple
WorkingDirectory=%h/timmy/hermes-agent
ExecStart=%h/timmy/venv/bin/python %h/timmy/hermes-agent/run_agent.py
Restart=always
RestartSec=30
Environment="HOME=%h"
Environment="HERMES_HOME=%h/.hermes"
[Install]
WantedBy=default.target

View File

@@ -0,0 +1,16 @@
[Unit]
Description=Hermes Laptop Daylight Agent
After=network.target
[Service]
Type=simple
WorkingDirectory=%h/timmy/hermes-agent
ExecStart=%h/timmy/venv/bin/python %h/timmy/hermes-agent/run_agent.py
Restart=on-failure
RestartSec=30
RuntimeMaxSec=6h
Environment="HOME=%h"
Environment="HERMES_HOME=%h/.hermes"
[Install]
WantedBy=default.target

View File

@@ -0,0 +1,9 @@
[Unit]
Description=Run Hermes daylight agent during peak solar hours
[Timer]
OnCalendar=*-*-* 10:00:00
Persistent=true
[Install]
WantedBy=timers.target

View File

@@ -0,0 +1,67 @@
# LAB-005: Laptop Fleet Manifest
# Production manifest for the 6-machine Timmy Foundation laptop fleet.
# Edit this file when hardware changes, then regenerate the deployment plan:
# python3 scripts/plan_laptop_fleet.py configs/laptop-fleet-manifest.yaml --markdown > docs/LAB-005-laptop-fleet-deployment.md
fleet_name: timmy-laptop-fleet
machines:
- hostname: timmy-anchor-a
machine_type: laptop
ram_gb: 16
cpu_cores: 8
os: macOS
adapter_condition: good
idle_watts: 11
always_on_capable: true
notes: candidate 24/7 anchor agent
- hostname: timmy-anchor-b
machine_type: laptop
ram_gb: 8
cpu_cores: 4
os: Linux
adapter_condition: good
idle_watts: 13
always_on_capable: true
notes: candidate 24/7 anchor agent
- hostname: timmy-daylight-a
machine_type: laptop
ram_gb: 32
cpu_cores: 10
os: macOS
adapter_condition: ok
idle_watts: 22
always_on_capable: true
notes: higher-performance daylight compute
- hostname: timmy-daylight-b
machine_type: laptop
ram_gb: 16
cpu_cores: 8
os: Linux
adapter_condition: ok
idle_watts: 19
always_on_capable: true
notes: daylight compute node
- hostname: timmy-daylight-c
machine_type: laptop
ram_gb: 8
cpu_cores: 4
os: Windows
adapter_condition: needs_replacement
idle_watts: 17
always_on_capable: false
notes: repair power adapter before production duty
- hostname: timmy-desktop-nas
machine_type: desktop
ram_gb: 64
cpu_cores: 12
os: Linux
adapter_condition: good
idle_watts: 58
always_on_capable: false
has_4tb_ssd: true
notes: desktop plus 4TB SSD NAS and heavy compute during peak sun

View File

@@ -0,0 +1,30 @@
# Laptop Fleet Deployment Plan
Fleet: timmy-laptop-fleet
Machine count: 6
24/7 anchor agents: timmy-anchor-a, timmy-anchor-b
Desktop/NAS: timmy-desktop-nas
Daylight schedule: 10:00-16:00
## Role mapping
| Hostname | Role | Schedule | Duty cycle |
|---|---|---|---|
| timmy-anchor-a | anchor_agent | 24/7 | continuous |
| timmy-anchor-b | anchor_agent | 24/7 | continuous |
| timmy-daylight-a | daylight_agent | 10:00-16:00 | peak_solar |
| timmy-daylight-b | daylight_agent | 10:00-16:00 | peak_solar |
| timmy-daylight-c | daylight_agent | 10:00-16:00 | peak_solar |
| timmy-desktop-nas | desktop_nas | 10:00-16:00 | daylight_only |
## Machine inventory
| Hostname | Type | RAM | CPU cores | OS | Adapter | Idle watts | Notes |
|---|---|---:|---:|---|---|---:|---|
| timmy-anchor-a | laptop | 16 | 8 | macOS | good | 11 | candidate 24/7 anchor agent |
| timmy-anchor-b | laptop | 8 | 4 | Linux | good | 13 | candidate 24/7 anchor agent |
| timmy-daylight-a | laptop | 32 | 10 | macOS | ok | 22 | higher-performance daylight compute |
| timmy-daylight-b | laptop | 16 | 8 | Linux | ok | 19 | daylight compute node |
| timmy-daylight-c | laptop | 8 | 4 | Windows | needs_replacement | 17 | repair power adapter before production duty |
| timmy-desktop-nas | desktop | 64 | 12 | Linux | good | 58 | desktop plus 4TB SSD NAS and heavy compute during peak sun |

View File

@@ -427,20 +427,6 @@ class ActionSystem:
"description": "Move to an adjacent room",
"target": "room",
},
"run": {
"cost": 3,
"description": "Run to an adjacent room (faster but more tiring)",
"target": "room",
},
"jump": {
"cost": 1,
"description": "Jump in place (brief energy surge)",
},
"gallop": {
"cost": 4,
"description": "Gallop across open ground (requires Threshold or Bridge)",
"target": "room",
},
"speak": {
"cost": 1,
"description": "Say something to someone in the room",
@@ -516,13 +502,6 @@ class ActionSystem:
connections = world.rooms[room].get("connections", {})
for direction, dest in connections.items():
available.append(f"move:{direction} -> {dest}")
available.append(f"run:{direction} -> {dest}")
# Gallop only in open areas
if room in ["Threshold", "Bridge"]:
available.append(f"gallop:{direction} -> {dest}")
# Jump always available
available.append("jump")
# Speaking (if others are here)
here = [n for n in world.characters if world.characters[n]["room"] == room and n != char_name]
@@ -1112,16 +1091,14 @@ class GameEngine:
"npc_actions": [],
"choices": [],
"log": [],
"particles": [],
}
# Process Timmy's action
timmy_energy = self.world.characters["Timmy"]["energy"]
room_name = self.world.characters["Timmy"]["room"] # current room for early exit messages
# Energy constraint checks
action_costs = {
"move": 2, "run": 3, "jump": 1, "gallop": 4, "tend_fire": 3, "write_rule": 2, "carve": 2,
"move": 2, "tend_fire": 3, "write_rule": 2, "carve": 2,
"plant": 2, "study": 2, "forge": 3, "help": 2, "speak": 1,
"listen": 0, "rest": -2, "examine": 0, "give": 0, "take": 1,
}
@@ -1182,10 +1159,6 @@ class GameEngine:
self.world.characters["Timmy"]["room"] = dest
self.world.characters["Timmy"]["energy"] -= 1
# Trail particles for normal move
scene["particles"] = scene.get("particles", [])
scene["particles"].append("✨ dust sparkles in your wake")
scene["log"].append(f"You move {direction} to The {dest}.")
scene["timmy_room"] = dest
@@ -1196,7 +1169,7 @@ class GameEngine:
# Check trust changes for arrival
here = [n for n in self.world.characters if self.world.characters[n]["room"] == dest and n != "Timmy"]
if here:
scene["log"].append(f"{repr(', ').join(here)} {('are' if len(here)>1 else 'is')} already here.")
scene["log"].append(f"{', '.join(here)} {'are' if len(here)>1 else 'is'} already here.")
for person in here:
self.world.characters[person]["trust"]["Timmy"] = min(1.0,
self.world.characters[person]["trust"].get("Timmy", 0) + 0.05)
@@ -1241,55 +1214,6 @@ class GameEngine:
else:
scene["log"].append("You can't go that way.")
elif timmy_action.startswith("run:"):
direction = timmy_action.split(":")[1]
current_room = self.world.characters["Timmy"]["room"]
connections = self.world.rooms[current_room].get("connections", {})
if direction in connections:
dest = connections[direction]
self.world.characters["Timmy"]["room"] = dest
self.world.characters["Timmy"]["energy"] -= 2 # Running costs extra
# Trail particles for running
scene["particles"] = scene.get("particles", [])
scene["particles"].append("💨 sprinting dust trail")
scene["particles"].append("✨ sparkles streaking")
scene["log"].append(f"You sprint {direction} to The {dest}.")
else:
scene["log"].append("You can't go that way.")
elif timmy_action == "jump":
# Jump in place - energizing
self.world.characters["Timmy"]["energy"] = min(10, self.world.characters["Timmy"]["energy"] + 0.5)
scene["particles"] = scene.get("particles", [])
scene["particles"].append("⭐ jump sparkle burst")
scene["log"].append("You jump! A burst of energy lifts you.")
elif timmy_action.startswith("gallop:"):
direction = timmy_action.split(":")[1]
current_room = self.world.characters["Timmy"]["room"]
connections = self.world.rooms[current_room].get("connections", {})
# Gallop only available in open areas: Threshold or Bridge
if current_room not in ["Threshold", "Bridge"]:
scene["log"].append("You need open ground to gallop. Find the Threshold or Bridge.")
elif direction in connections:
dest = connections[direction]
self.world.characters["Timmy"]["room"] = dest
self.world.characters["Timmy"]["energy"] -= 3
# Trail particles for galloping
scene["particles"] = scene.get("particles", [])
scene["particles"].append("🌟 galloping star dust")
scene["particles"].append("✨ running sparkles")
scene["particles"].append("💨 wind trail")
scene["log"].append(f"You gallop {direction} to The {dest}. Hooves thunder.")
else:
scene["log"].append("You can't gallop that way.")
elif timmy_action.startswith("speak:"):
target = timmy_action.split(":")[1]
if self.world.characters[target]["room"] == self.world.characters["Timmy"]["room"]:

View File

@@ -1,12 +1 @@
# Timmy core module
from .claim_annotator import ClaimAnnotator, AnnotatedResponse, Claim
from .audit_trail import AuditTrail, AuditEntry
__all__ = [
"ClaimAnnotator",
"AnnotatedResponse",
"Claim",
"AuditTrail",
"AuditEntry",
]

View File

@@ -1,156 +0,0 @@
#!/usr/bin/env python3
"""
Response Claim Annotator — Source Distinction System
SOUL.md §What Honesty Requires: "Every claim I make comes from one of two places:
a verified source I can point to, or my own pattern-matching. My user must be
able to tell which is which."
"""
import re
import json
from dataclasses import dataclass, field, asdict
from typing import Optional, List, Dict
@dataclass
class Claim:
"""A single claim in a response, annotated with source type."""
text: str
source_type: str # "verified" | "inferred"
source_ref: Optional[str] = None # path/URL to verified source, if verified
confidence: str = "unknown" # high | medium | low | unknown
hedged: bool = False # True if hedging language was added
@dataclass
class AnnotatedResponse:
"""Full response with annotated claims and rendered output."""
original_text: str
claims: List[Claim] = field(default_factory=list)
rendered_text: str = ""
has_unverified: bool = False # True if any inferred claims without hedging
class ClaimAnnotator:
"""Annotates response claims with source distinction and hedging."""
# Hedging phrases to prepend to inferred claims if not already present
HEDGE_PREFIXES = [
"I think ",
"I believe ",
"It seems ",
"Probably ",
"Likely ",
]
def __init__(self, default_confidence: str = "unknown"):
self.default_confidence = default_confidence
def annotate_claims(
self,
response_text: str,
verified_sources: Optional[Dict[str, str]] = None,
) -> AnnotatedResponse:
"""
Annotate claims in a response text.
Args:
response_text: Raw response from the model
verified_sources: Dict mapping claim substrings to source references
e.g. {"Paris is the capital of France": "https://en.wikipedia.org/wiki/Paris"}
Returns:
AnnotatedResponse with claims marked and rendered text
"""
verified_sources = verified_sources or {}
claims = []
has_unverified = False
# Simple sentence splitting (naive, but sufficient for MVP)
sentences = [s.strip() for s in re.split(r'[.!?]\s+', response_text) if s.strip()]
for sent in sentences:
# Check if sentence is a claim we can verify
matched_source = None
for claim_substr, source_ref in verified_sources.items():
if claim_substr.lower() in sent.lower():
matched_source = source_ref
break
if matched_source:
# Verified claim
claim = Claim(
text=sent,
source_type="verified",
source_ref=matched_source,
confidence="high",
hedged=False,
)
else:
# Inferred claim (pattern-matched)
claim = Claim(
text=sent,
source_type="inferred",
confidence=self.default_confidence,
hedged=self._has_hedge(sent),
)
if not claim.hedged:
has_unverified = True
claims.append(claim)
# Render the annotated response
rendered = self._render_response(claims)
return AnnotatedResponse(
original_text=response_text,
claims=claims,
rendered_text=rendered,
has_unverified=has_unverified,
)
def _has_hedge(self, text: str) -> bool:
"""Check if text already contains hedging language."""
text_lower = text.lower()
for prefix in self.HEDGE_PREFIXES:
if text_lower.startswith(prefix.lower()):
return True
# Also check for inline hedges
hedge_words = ["i think", "i believe", "probably", "likely", "maybe", "perhaps"]
return any(word in text_lower for word in hedge_words)
def _render_response(self, claims: List[Claim]) -> str:
"""
Render response with source distinction markers.
Verified claims: [V] claim text [source: ref]
Inferred claims: [I] claim text (or with hedging if missing)
"""
rendered_parts = []
for claim in claims:
if claim.source_type == "verified":
part = f"[V] {claim.text}"
if claim.source_ref:
part += f" [source: {claim.source_ref}]"
else: # inferred
if not claim.hedged:
# Add hedging if missing
hedged_text = f"I think {claim.text[0].lower()}{claim.text[1:]}" if claim.text else claim.text
part = f"[I] {hedged_text}"
else:
part = f"[I] {claim.text}"
rendered_parts.append(part)
return " ".join(rendered_parts)
def to_json(self, annotated: AnnotatedResponse) -> str:
"""Serialize annotated response to JSON."""
return json.dumps(
{
"original_text": annotated.original_text,
"rendered_text": annotated.rendered_text,
"has_unverified": annotated.has_unverified,
"claims": [asdict(c) for c in annotated.claims],
},
indent=2,
ensure_ascii=False,
)

View File

@@ -50,3 +50,43 @@ def test_manifest_template_is_valid_yaml() -> None:
data = yaml.safe_load(Path("docs/laptop-fleet-manifest.example.yaml").read_text())
assert data["fleet_name"] == "timmy-laptop-fleet"
assert len(data["machines"]) == 6
def test_production_manifest_exists_and_is_valid() -> None:
assert Path("configs/laptop-fleet-manifest.yaml").exists()
data = yaml.safe_load(Path("configs/laptop-fleet-manifest.yaml").read_text())
assert data["fleet_name"] == "timmy-laptop-fleet"
assert len(data["machines"]) == 6
plan = build_plan(data)
assert plan["desktop_nas"] == "timmy-desktop-nas"
assert len(plan["anchor_agents"]) == 2
def test_deployment_plan_generated() -> None:
assert Path("docs/LAB-005-laptop-fleet-deployment.md").exists()
content = Path("docs/LAB-005-laptop-fleet-deployment.md").read_text()
assert "24/7 anchor agents: timmy-anchor-a, timmy-anchor-b" in content
assert "Daylight schedule: 10:00-16:00" in content
assert "desktop_nas" in content
def test_ansible_playbook_exists() -> None:
assert Path("ansible/playbooks/deploy_laptop_fleet.yml").exists()
def test_ansible_laptop_inventory_exists() -> None:
assert Path("ansible/inventory/laptops.ini").exists()
content = Path("ansible/inventory/laptops.ini").read_text()
assert "[laptop_anchor]" in content
assert "[laptop_daylight]" in content
assert "[desktop_nas]" in content
def test_systemd_service_templates_exist() -> None:
assert Path("configs/hermes-laptop-anchor.service").exists()
assert Path("configs/hermes-laptop-daylight.service").exists()
assert Path("configs/hermes-laptop-daylight.timer").exists()
anchor = Path("configs/hermes-laptop-anchor.service").read_text()
daylight = Path("configs/hermes-laptop-daylight.service").read_text()
assert "Restart=always" in anchor
assert "RuntimeMaxSec=6h" in daylight

View File

@@ -1,103 +0,0 @@
#!/usr/bin/env python3
"""Tests for claim_annotator.py — verifies source distinction is present."""
import sys
import os
import json
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "src"))
from timmy.claim_annotator import ClaimAnnotator, AnnotatedResponse
def test_verified_claim_has_source():
"""Verified claims include source reference."""
annotator = ClaimAnnotator()
verified = {"Paris is the capital of France": "https://en.wikipedia.org/wiki/Paris"}
response = "Paris is the capital of France. It is a beautiful city."
result = annotator.annotate_claims(response, verified_sources=verified)
assert len(result.claims) > 0
verified_claims = [c for c in result.claims if c.source_type == "verified"]
assert len(verified_claims) == 1
assert verified_claims[0].source_ref == "https://en.wikipedia.org/wiki/Paris"
assert "[V]" in result.rendered_text
assert "[source:" in result.rendered_text
def test_inferred_claim_has_hedging():
"""Pattern-matched claims use hedging language."""
annotator = ClaimAnnotator()
response = "The weather is nice today. It might rain tomorrow."
result = annotator.annotate_claims(response)
inferred_claims = [c for c in result.claims if c.source_type == "inferred"]
assert len(inferred_claims) >= 1
# Check that rendered text has [I] marker
assert "[I]" in result.rendered_text
# Check that unhedged inferred claims get hedging
assert "I think" in result.rendered_text or "I believe" in result.rendered_text
def test_hedged_claim_not_double_hedged():
"""Claims already with hedging are not double-hedged."""
annotator = ClaimAnnotator()
response = "I think the sky is blue. It is a nice day."
result = annotator.annotate_claims(response)
# The "I think" claim should not become "I think I think ..."
assert "I think I think" not in result.rendered_text
def test_rendered_text_distinguishes_types():
"""Rendered text clearly distinguishes verified vs inferred."""
annotator = ClaimAnnotator()
verified = {"Earth is round": "https://science.org/earth"}
response = "Earth is round. Stars are far away."
result = annotator.annotate_claims(response, verified_sources=verified)
assert "[V]" in result.rendered_text # verified marker
assert "[I]" in result.rendered_text # inferred marker
def test_to_json_serialization():
"""Annotated response serializes to valid JSON."""
annotator = ClaimAnnotator()
response = "Test claim."
result = annotator.annotate_claims(response)
json_str = annotator.to_json(result)
parsed = json.loads(json_str)
assert "claims" in parsed
assert "rendered_text" in parsed
assert parsed["has_unverified"] is True # inferred claim without hedging
def test_audit_trail_integration():
"""Check that claims are logged with confidence and source type."""
# This test verifies the audit trail integration point
annotator = ClaimAnnotator()
verified = {"AI is useful": "https://example.com/ai"}
response = "AI is useful. It can help with tasks."
result = annotator.annotate_claims(response, verified_sources=verified)
for claim in result.claims:
assert claim.source_type in ("verified", "inferred")
assert claim.confidence in ("high", "medium", "low", "unknown")
if claim.source_type == "verified":
assert claim.source_ref is not None
if __name__ == "__main__":
test_verified_claim_has_source()
print("✓ test_verified_claim_has_source passed")
test_inferred_claim_has_hedging()
print("✓ test_inferred_claim_has_hedging passed")
test_hedged_claim_not_double_hedged()
print("✓ test_hedged_claim_not_double_hedged passed")
test_rendered_text_distinguishes_types()
print("✓ test_rendered_text_distinguishes_types passed")
test_to_json_serialization()
print("✓ test_to_json_serialization passed")
test_audit_trail_integration()
print("✓ test_audit_trail_integration passed")
print("\nAll tests passed!")