peregrine/config/llm.cloud.yaml
pyr0ball 8e36863a49
Some checks failed
CI / Backend (Python) (push) Failing after 2m15s
CI / Frontend (Vue) (push) Failing after 21s
Mirror / mirror (push) Failing after 9s
feat: Interview prep Q&A, cf-orch hardware profile, a11y fixes, dark theme
Backend
- dev-api.py: Q&A suggest endpoint, Log Contact, cf-orch node detection in wizard
  hardware step, canonical search_profiles format (profiles:[...]), connections
  settings endpoints, Resume Library endpoints
- db_migrate.py: migrations 002/003/004 — ATS columns, resume review, final
  resume struct
- discover.py: _normalize_profiles() for legacy wizard YAML format compat
- resume_optimizer.py: section-by-section resume parsing + scoring
- task_runner.py: Q&A and contact-log task types
- company_research.py: accessibility brief column wiring
- generate_cover_letter.py: restore _candidate module-level binding

Frontend
- InterviewPrepView.vue: Q&A chat tab, Log Contact form, MarkdownView rendering
- InterviewCard.vue: new reusable card component for interviews kanban
- InterviewsView.vue: rejected analytics section with stage breakdown chips
- ResumeProfileView.vue: sync with new resume store shape
- SearchPrefsView.vue: cf-orch toggle, profile format migration
- SystemSettingsView.vue: connections settings wiring
- ConnectionsSettingsView.vue: new view for integration connections
- MarkdownView.vue: new component for safe markdown rendering
- ApplyWorkspace.vue: a11y — h1→h2 demotion, aria-expanded on Q&A toggle,
  confirmation dialog on Reject action (#98 #99 #100)
- peregrine.css: explicit [data-theme="dark"] token block for light-OS users (#101),
  :focus-visible outline (#97)
- wizard.css: cf-orch hardware step styles
- WizardHardwareStep.vue: cf-orch node display, profile selection with orch option
- WizardLayout.vue: hardware step wiring

Infra
- compose.yml / compose.cloud.yml: cf-orch agent sidecar, llm.cloud.yaml mount
- Dockerfile.cfcore: cf-core editable install in image build
- HANDOFF-xanderland.md: Podman/systemd setup guide for beta tester
- podman-standalone.sh: standalone Podman run script

Tests
- test_dev_api_settings.py: remove stale worktree path bootstrap (credential_store
  now in main repo); fix job_boards fixture to use non-empty list
- test_wizard_api.py: update profiles assertion to superset check (cf-orch added);
  update step6 assertion to canonical profiles[].titles format
2026-04-14 17:01:18 -07:00

72 lines
1.7 KiB
YAML

backends:
anthropic:
api_key_env: ANTHROPIC_API_KEY
enabled: false
model: claude-sonnet-4-6
supports_images: true
type: anthropic
claude_code:
api_key: any
base_url: http://localhost:3009/v1
enabled: false
model: claude-code-terminal
supports_images: true
type: openai_compat
github_copilot:
api_key: any
base_url: http://localhost:3010/v1
enabled: false
model: gpt-4o
supports_images: false
type: openai_compat
ollama:
api_key: ollama
base_url: http://host.docker.internal:11434/v1
enabled: true
model: llama3.1:8b # generic — no personal fine-tunes in cloud
supports_images: false
type: openai_compat
ollama_research:
api_key: ollama
base_url: http://host.docker.internal:11434/v1
enabled: true
model: llama3.1:8b
supports_images: false
type: openai_compat
vision_service:
base_url: http://host.docker.internal:8002
enabled: true
supports_images: true
type: vision_service
vllm:
api_key: ''
base_url: http://host.docker.internal:8000/v1
enabled: true
model: __auto__
supports_images: false
type: openai_compat
cf_orch:
service: vllm
model_candidates:
- Qwen2.5-3B-Instruct
ttl_s: 300
vllm_research:
api_key: ''
base_url: http://host.docker.internal:8000/v1
enabled: true
model: __auto__
supports_images: false
type: openai_compat
cf_orch:
service: vllm
model_candidates:
- Qwen2.5-3B-Instruct
ttl_s: 300
fallback_order:
- vllm
- ollama
research_fallback_order:
- vllm_research
- ollama_research
vision_fallback_order:
- vision_service