Drop stale gemini_cli expectations: PROVIDERS expectations in
test_providers_source_of_truth + test_provider_curation, the
test_list_models_gemini_cli function (deleted the whole file since it
was the only test), the test_discussion_compression_gemini_cli function,
the test_gcli_path_updates_adapter function (its setter was removed),
the ui_gemini_cli_path references in test_mma_tier_usage_reset_fix and
test_rag_integration. These are the t1.9 minor edits.
phase_1.status = in_progress_partial_6_of_11; tasks 1.1-1.6 marked completed with SHAs. Tasks 1.7 (mock provider) through 1.11 (batch run) remain pending. Phases 2 and 3 entirely untouched. current_phase advanced from 0 to 1 to reflect Front A mid-execution.
Tasks 1.1-1.6 cover the bulk of Front A: the gemini_cli adapter module
plus all importer edits (ai_client, app_controller, gui_2,
project_manager, api_hooks) and the 7 dedicated gemini_cli test files.
Remaining Front A tasks (1.7-1.11) cover the mock provider, the 11 sim
test rewrites, the minor test edits, the doc updates, and the batch
checkpoint.
Drop the 11 gemini_cli sites: the ui_gemini_cli_path state field, the
gcli_path settable map entry, the _update_gcli_adapter method, the
project load/save gemini_cli binary_path entries, and the per-provider
gemini_cli guard at startup.
Drop the standalone Gemini CLI adapter from the AI client surface: delete
the import, the PROVIDERS entry, the module state, the 3 functions
(_list_gemini_cli_models, _send_cli_round_result, _send_gemini_cli), and
the 8 dispatch branches. PROVIDERS now has 7 entries; _gemini_sdk
remainder is unaffected.
3 fronts, 3 phases, 31 tasks, 12 verification criteria. Based on 3
parallel explore-agent audits of the live codebase.
Front A — Gemini CLI Adapter Removal: delete src/gemini_cli_adapter.py
+ 7 test files; edit ai_client.py (3 functions + 8 dispatch branches +
PROVIDERS 8->7); edit app_controller.py (~11 sites) + gui_2.py (2 blocks)
+ project_manager.py + api_hooks.py; introduce a mock provider that
routes to tests/mock_gemini_cli.py so the 9 live_gui sim tests switch
from gemini_cli to mock without losing their mock-LLM mechanism; update
7 docs + 5 conductor docs.
Front B — Test Suite Cleanup (410 files, 2,133 tests, ~45-60 files
downsizable): delete 2 guaranteed-broken chronology tests (import from
deleted scripts.audit.*); consolidate 7 test_scavenge_*.py tautologies
into 1 test_directive_structure.py; review deprecated-module tests;
consolidate ~15 test_*_phase*.py one-offs; migrate 27 src.models shim
importers to direct subsystem imports; extract shared mock_controller
fixture from ~8 boilerplate-patch files; fix 3 fix-not-skip candidates
(Gemini 503 in summarize.summarise_file); audit the 120KB
test_gui_2_result.py (9.4% of suite line count in one file).
Front C — Metadata-Type Reduction: flip the 38-site MCP dispatch
inversion (dispatch()+22 consumers still call str wrappers, not the
_result variants); migrate app_controller.py's ~40 in-memory Metadata
state sites to typed per-aggregate dataclasses; fix aggregate.py
file_items: list[Metadata] -> list[FileItem]; delete dead code
(openai_schemas.to_legacy_dict, app_controller._push_mma_state_update).
Out of scope: GEMINI_CLI_HOOK_CONTEXT env var + cli_tool_bridge.py +
mma_exec.py (meta-tooling, not the provider — separate follow-up if
user wants to purge naming).
download_media now shells out to gallery-dl (handles X.com auth/403/video->mp4) instead of urllib, returns the downloaded files, + media_files_for_post glob helper + --cookies CLI. render_markdown embeds images with  and videos with an HTML <video controls> tag (were plain links). extract_corpus uses the package download_media (one download path). Tests rewritten for the gallery-dl API. Corpus: 4 threads, 33 media (32 img embeds + 1 video), 0 missing.
Replaced convert_cookies.py + run_corpus.py + dedupe_corpus.py (+ the scratch merge_corpus.py) with a single extract_corpus.py that: converts cookies, fetches each URL's full conversation, MERGES overlapping conversations into one thread (union of posts, no duplicate threads/media), downloads ALL media (images + video/mp4) via gallery-dl itself (plain urllib 403'd on pbs.twimg.com and never fetched videos), and renders. Result: 4 merged threads, 33 media files (incl mp4), 0 missing links. cookies_netscape.txt gitignored.
fetch_thread_from_url now passes -o conversations=true (full threads, not single tweets), sorts posts chronologically, and sets root_post_id to the URL's target tweet. render_markdown builds front-matter from the target (root_post_id) post, not the earliest conversation tweet (fixes wrong @handle). Added dedupe_corpus.py: keeps deepest thread per conversation, deletes duplicates, keeps+marks unique branches (branch_of) + writes threads_index.json. 33 tests pass.
Script used calendar-day duration (end_date - init_date), which
inflated tracks that sat dormant between work and archive-move.
Example: external_editor_integration_20260308 reported 60 days
(init 2026-03-08, last archive-move 2026-05-07) when actual work
was 2 active days.
Corrected metric: active days = distinct calendar days with at
least one commit touching the track folder. All 30 top tracks
now show 2-7 active days, matching the user's expectation that
no track lasts more than a week.
Also fixed commit_count + files_touched for all 30 tracks (the
script's work-prefix-only count was always 0-1 for these).
Top 3 examples:
- external_editor_integration_20260308: 60d/1c/4f -> 2d/4c/8f
- code_path_audit_20260607: 18d/1c/7f -> 7d/23c/14f
- nagent_review_20260608: 12d/0c/20f -> 3d/53c/20f
Still stale (need recompute, not hand-fix):
- 3.1 per-era aggregates
- 8 use-case predictions
- 218 other tracks not in top-30
The script's commit_count and files_touched columns only counted
work-prefix commits (feat|fix|refactor|perf|test) with pathspecs
matching the track folder. Track folders contain only metadata
commits (conductor(*), chore(conductor), docs(track)), so the
script reported 0 commits for any track that did real work.
Corrected rows (now using total commits + total unique files
touching the track folder, any prefix):
- code_path_audit_20260607: 1->23 commits, 7->14 files
- qwen_llama_grok_integration_20260606: 0->25 commits, 4->8 files
- nagent_review_20260608: 0->53 commits, 4->20 files
- data_oriented_error_handling_20260606: 0->21 commits, 4->8 files
- superpowers_review_20260619: 0->63 commits, 8->16 files
Added 9.1 caveat explaining the manual corrections and the
methodology error that caused them.
thread_from_dict (JSON wire boundary) round-trips thread_data.json; download_media/render_markdown gain argparse main() + __main__ so the plan Task 5.3 pipeline runs via -m; dual-import added so all modules run standalone or via -m (G5). Pipeline integration test (html->names->render) + CLI tests cover the glue without network. TDD: red -> green (4 pipeline + 29 regression = 33 passed).
- Move conductor/tracks/meta_tooling_duration_analysis_20260705/ to
conductor/archive/meta_tooling_duration_analysis_20260705/ (preserves
history as a directory rename since the spec/plan/metadata files were
uncommitted; they're added to the archive in this commit).
- Prepend the track's row to conductor/chronology.md (newest-first).
- Note: spec.md, plan.md, metadata.json, state.toml are being committed
for the first time in this archive commit. Per project convention,
spec + plan are committed at archive time per workflow.md chronology
maintenance rules.
Added standalone Result[T] (frozen+slots, ok/err classmethods, is_ok) to error_types.py per canonical AND-over-OR error_handling.md (not the video_analysis _Ok|_Err sum type). render_markdown renders YAML front-matter from root post, per-post ## sections with reply markers, quote blockquotes, and ./media/<name> links; media_names mapping keeps render decoupled from download_media. TDD: red (no Result) -> green (7 render + 9 types passed).
The original build_track_index spawned 5+ git subprocesses per track
folder (244 tracks * 5 calls = 1200+ subprocesses). That exceeded the
120s NFR1 budget. Replaced with a single 'git log --all --name-only'
pass + a mutable _TrackAccumulator; reduces git subprocess count from
1200+ to 1.
Also adds 3 smoke tests for build_subagent_task_index,
build_session_index, build_track_index (covering previously untested
pure-function indexers). 23 tests pass; coverage up from 39% to 61%.
Scripts + workflow for extracting Twitter/X posts and threads into
Markdown with associated media. Mirrors the scripts/video_analysis/
pattern. Standalone requirement: zero imports from src/, conductor/,
or scripts.video_analysis — copy-pasteable to another repo with only
gallery-dl as the external dep.
5 modules: __init__.py, error_types.py (Result[T, ErrorInfo] +
ThreadData/PostData typed dataclasses), fetch_thread.py (gallery-dl
subprocess for URLs + html.parser fallback for local HTML),
download_media.py (stdlib urllib, idempotent), render_markdown.py
(YAML front-matter + per-post sections + ./media/ links).
Reference project: C:\projects\forth\bootslop — the corpus feeds
bootslop's scripts and reference-generation pipeline. Acceptance
corpus: 8 threads (@NOTimothyLottes x6 + @VPCOMPRESSB x2) extracted
to tests/artifacts/twitter_threads_corpus/. The ?s=20 quote-share
suffix on the @VPCOMPRESSB URLs must be stripped by fetch_thread.py
before acquisition (added to FR2 as URL normalization).
5 phases / 23 tasks. 8 verification criteria (VC1-VC8). TDD red-first
on the pure-function modules (render_markdown, types, media naming).
Deleted scripts/audit/generate_chronology.py + chronology_quality_gate.py
after repeated corruption incidents (auto-classifier drifted from
user intent, silently rewrote rows). chronology.md is now maintained
by hand: add a row at the top when a track ships/abandons/is archived.
Updated conductor/workflow.md §Chronology Maintenance and
conductor/tracks.md §Archiving a track to reflect manual maintenance.
Added 40 rows to conductor/chronology.md for the 2026-07-05 archive
batch, with archive-folder paths and deferral notes where applicable.
Removed 12 archived-track rows from conductor/tracks.md Active Tracks
table (rows 2, 3, 7c, 16, 17, 23b, 23c, 23d, 25, 26, 29, 29c) and
their track-detail anchors.
Wrote docs/reports/ARCHIVE_REVIEW_20260705.md: a categorized index
of unfinished work items from the archived tracks' state.toml files
that remain relevant to the codebase, with a ranked follow-up list.
Per agent_directives_consolidation_20260705 §3.4.
The 14-line §Data Structure Conventions section (the 'names for shapes'
pattern intro + the 16-alias list + canonical reference) is replaced
with a 1-line pointer to conductor/code_styleguides/type_aliases.md,
the canonical home for the 10 TypeAlias definitions + 11 per-aggregate
dataclasses + 5-pattern decision tree.
Net: 14 lines reduced to 1 line of pointer. The type-registry
auto-generation note (docs/type_registry/) is preserved as unique
project information.
Phase 3.2, 3.3, 3.4 all complete in this single commit batch.
ed