- fl.attorney.agent (AbstractModel): manual-only substantive analysis fired from
the case AI tab. Builds a full case context (parties, children, financials,
issue tags, prior analyses) plus candidate statute/caselaw lists, and asks
Claude to author a strategy memo, draft arguments/counterarguments, write a
risk narrative, and assess substantial change (FL 61.30(1)(b))
- Grounds output in the real library: the model may only pick statutes/case law
from the supplied candidates, which are then resolved back to records and
linked (fl.analysis.cited_statute_ids / matched_caselaw_ids, case.caselaw_ids)
- Rule-based fallback produces a usable memo (complexity, statutes by category,
caselaw by tag, risk flags) when the API is unavailable — never a raw error
- fl.analysis: add analysis_type, strategy_memo (Html), risk_narrative,
cited_statute_ids; surface them in the analysis views
- fl.case: add attorney_memo_id + related memo/risk display; action_run_attorney_agent
opens the memo; "Generate Attorney Strategy Memo" button on the AI tab (admin)
- Refactor fl_ai_engine: extract shared call_claude_json(system, user) +
_extract_json so both agents and the engine share one Claude/JSON path
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Stage alignment:
- Replace the 11 procedural stages with the spec's 5-stage machine
(Intake/Active/Discovery/Pre-Trial/Closed); only Closed is folded
- Make fl_stage_data.xml updatable (drop noupdate) so the rename applies on
upgrade; keep fl_stage_intake/discovery/closed XML ids
- Update case search filters to the new stage names (Intake/Active/Discovery/Pre-Trial)
Issue-tag bug fix (fl_ai_engine):
- Rule-based tagging and caselaw matching compared snake_case keys against the
human-readable seeded tag names, so they never matched and issue_tag_ids was
never populated. Add RULE_KEY_TO_TAG_NAME and translate keys to real names
before all fl.issue.tag / fl.caselaw lookups
Paralegal agent (fl.paralegal.agent, AbstractModel):
- on_stage_change(): fast rule-based pass fired automatically on stage entry and
case creation — generates the stage task batch (idempotent), recalculates
filing/service deadlines, cross-references statutes by issue tag + case type,
posts a chatter summary. No Claude call, so it never blocks the workflow
- run_manual(): full pass adding a best-effort Claude procedural briefing with
rule-based fallback; wired to a "Paralegal Review" button on the case form
- AI audit-time logging is guarded behind a fl.timesheet existence check
(model not built yet)
- fl.case.write fires on_stage_change only when stage_id actually changes;
create() generates the Intake batch
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Replace fl.case.stage Selection field with Many2one → fl.case.stage model,
enabling Kanban grouping and dynamic stage management
- Add FlCaseStage model (sequence, fold, description) and fl_stage_data.xml
with all 11 procedural stages seeded with noupdate=1
- Migrate fl_ai_engine.py from Ollama/llama3.1 to Claude API
(claude-sonnet-4-20250514); key from ir.config_parameter fl_ai.claude_api_key
- Fix stale field references in _rule_based_tagging and _build_case_context:
employment/income now read from party_ids, timesharing fields corrected
- Add _fallback_complexity() for graceful degradation when API unavailable
- Add Kanban view to fl_case_views.xml; update action view_mode to kanban,tree,form
- Add fl.case.stage ACL entries to ir.model.access.csv
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
