docs: rename app to Samaritan and sync docs#1
Merged
Conversation
Includes skills registry, engine view, signal view, career patterns, user skills API, and all frontend/backend changes to date.
Add migrations 0017-0022 to create normalized child tables for project_skills, project_tech_stack, project_target_roles, project_evidence, learning_concepts, learning_tools, learning_code_snippets, message_sources, and impression_candidates. Backfill existing JSON data during migration. Update all service-layer reads to query child tables and writes to dual-write (JSON + child tables) for safe incremental rollout. Store event large payloads inline in payload_text column instead of on disk.
Add migration 0023 to DROP the now-redundant JSON columns: skills_json, tech_stack_json, target_roles_json, evidence_json (projects), concepts_json, tools_json, code_snippets_json (learnings), sources_json (messages), candidates_json (impressions). Remove all json.dumps/json.loads for normalized columns from service layer. Remove import json from projects.py, extract_projects.py, and learnings_store.py. Remove _source_list_adapter and TypeAdapter import from history.py. Update _migrate_json_files to write directly to child tables. Update migration test to expect sources_json absent.
Cover the full skill routing → dispatch → response flow for both sync (process_chat) and streaming (process_chat_stream) paths, including slash command dispatch, fallback when skill is missing, skills list verification, and startup registration checks.
Wire up /api/skills endpoint and surface skill awareness in the chat UI: - Slash autocomplete dropdown when typing / in chat input with keyboard nav - Emerald "Skill: <name>" badge on responses with answer_origin: "skill" - "Analyzing..." status text for the new analyzing stream status
Snapshot of pattern impact preview work from parallel session.
scaleborg
added a commit
that referenced
this pull request
Mar 14, 2026
Add 3 cases (tc_fig_001-003) that query figure description content from the IRPAPERS paper. Tagged with "figure" for filtering. - tc_fig_001: numeric lookup (recall@20 value from bar chart) - tc_fig_002: method comparison (best at Recall@1) - tc_fig_003: pipeline architecture (diagram components) Results: Faithfulness 0.95-1.0, Context Recall 1.0 on all three. Figure chunks rank #1-3 in retrieval for all cases. Dataset version bumped to 2.1.0 (15 → 18 cases).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary\n- rename app naming in README/docs/Mintlify from Second Brain to Samaritan\n- align docs API/contracts with current product surfaces\n- include Mintlify contract/schema updates and cleanup of unused Mintlify images\n\nOpened to force CI on this branch.