Save each discovery conversation's prompts and answers to durable CSV
files (per-conversation + append-only master log) on both save and
completion, so answers survive an extraction error, can be re-fed to the
AI, and can be reviewed/resumed by the user.
- app/services/answer_store.py: canonical prompt list + atomic CSV writes,
master append, and read-back helpers (DB stays system of record; CSV
failures are logged, never fatal).
- discovery router: write CSV on /respond and /complete; new endpoints
GET /answers, GET /answers.csv, POST /reprocess (shared extraction
helper; locked profiles return 409).
- discovery.html: prefill/resume from saved answers after an error and a
"Re-run analysis" button wired to /reprocess.
- scripts/reprocess_csv.py: offline CLI to re-run extraction from a CSV
(print or --write-db).
- QUESTIONS_DIR / QUESTIONS_MASTER_CSV config, .gitignore, README, tests.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Dg6XWUwprmP5QCL18HxssY
Final roadmap phase. No DB migration — it reads data already captured.
- Goal-evolution history: GET /discovery/profile/me/goal-history derives a
per-goal timeline from the profile_revision snapshots (pure aggregator in
app/services/profile_history.py).
- Smart tagging: POST /discovery/integration/suggest-foundation suggests which
foundation a task builds toward + rationale/confidence (FoundationTagger,
app/services/tagging.py). Suggestion only; the person confirms by posting the
task mapping.
- Deeper goal-refinement: the reflect loop accepts an optional focus ("goals")
that steers the coach toward sharpening goals — still mirror, not compass.
- Visualizations: visuals.html renders an Ikigai Venn and an Enneagram diagram
(plain-language callouts, not the raw type number) plus the goal-evolution
timeline; linked from profile.html.
Tests: 99 passing (added pure goal-history tests, goal-history + suggest
endpoint tests, reflect-focus passthrough; run in-container). README updated.
This completes the ImpactFlow Vision roadmap (Phases 1-5) on the Discovery side.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add the mirror-not-compass reflection layer between profile generation and
affirmation. The coach reflects the person's profile back, and only when they
explicitly correct or add something does it propose revisions in their own
direction — never prescribing goals.
- ReflectionCoach service (app/services/reflector.py): Anthropic-backed,
returns {message, revisions, revision_note}; revisions filtered to the seven
editable prose fields (never triad/type); one-retry JSON handling.
- Endpoints (owner-scoped, 409 when locked): POST /discovery/profile/me/reflect
(opener + turns, applies revisions), GET .../reflection (dialogue),
GET .../revisions (iteration history). complete records an 'extraction'
revision; PATCH records 'manual_edit'.
- Models + migration 004: reflection_message (coach/person turns) and
profile_revision (snapshots: extraction | reflection | manual_edit) —
captures edits and iterations rather than overwriting.
- Frontend: reflect.html chat (coach/person bubbles, live profile summary that
refreshes on revision, affirm); linked from profile.html.
- Affirmation remains the existing confirm/lock.
Also refresh README for Phase 2 and for the HTTPS deployment
(https://impactflow.teamci.org:8011, OAUTH_REDIRECT_URI + COOKIE_SECURE notes).
Tests: 50 passing (added reflector unit tests and reflection endpoint tests;
run in-container).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Close the remaining Phase 1 DoD gaps and reconcile the browser flow with
the auth layer.
Goals (5 -> 7 prompts):
- Add near-term (6-12mo) and long-term (3-5yr) goal prompts; collect raw
text on the conversation and store AI-articulated goal summaries on the
profile. Extractor articulates the person's own stated goals (mirror,
not compass) and never fabricates. Alembic 003 adds the four columns.
Cookie-based browser sessions (fixes frontend<->auth desync):
- OAuth callback now sets httpOnly session cookies and redirects into the
app instead of returning JSON. get_current_user gains a cookie fallback
(X-API-Key -> Bearer -> cookie). refresh/logout read the refresh cookie
and set/clear cookies. New shared auth.js (authedFetch) sends cookies and
silently refreshes on 401. Static pages drop the bogus user_id and call
the correct /me endpoints.
Profile editing (read/edit/affirm):
- PATCH /discovery/profile/me edits the prose (Ikigai summaries, overlap
narrative, goals); owner-scoped, partial update, 409 when locked. Edit
mode in profile.html with Save/Cancel.
Also: bump default model to claude-sonnet-4-6, align ports to 8011
(OAuth redirect, CORS), add COOKIE_SECURE/POST_LOGIN_REDIRECT config, and
refresh the README to match the shipped behavior.
Tests: 33 passing (added cookie-auth, profile-edit, goal-extraction cases;
factored a shared app_client fixture into conftest.py).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude