coulomb/state-hub
2
0
Fork 0
generated from coulomb/repo-seed
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
354 Commits 1 Branch 0 Tags
18a5e2d6f0c41548c33cb5d6c2e9567a721a4e37
Commit Graph

6 Commits

Author SHA1 Message Date
tegwick
0949d4c0d8 feat(classification-spine): implement STATE-WP-0065 repo-anchored model 
Replace the ad-hoc coordination-domain spine with the Repo Classification
Standard: 14 market domains, classification columns on managed_repos, and
workplans anchored by repo_id (topic_id optional).

- Add Alembic migration d8e9f0a1b2c3 with data backfill and workstream→workplan rename
- Add api/classification.py validation and register-from-classification tooling
- Expose workplan-first REST/MCP surface with legacy workstream aliases
- Add C-24 consistency rule and legacy domain frontmatter mapping
- Update dashboard repos page with category/capability/stake filters
- Update orientation docs; mark STATE-WP-0065 finished
 2026-06-22 13:52:13 +02:00
tegwick
38835e9e79 feat(tasks): adopt canonical task statuses 2026-05-26 01:32:50 +02:00
tegwick
c12091c2eb Fixed and improved token tracking 2026-05-23 13:59:05 +02:00
tegwick
af3fdfde80 feat(token-tracking): introduce token note taxonomy (measured/userbased/workplan/heuristic) 
Tier 1 (exact counts) now defaults to note="measured" instead of null,
signalling the counts were read from the Claude Code status bar.
Callers can pass note="userbased" when a human provided the numbers.

  measured  — agent read exact counts from the Claude Code status bar
  userbased — counts provided by a human
  workplan  — prorated from workplan total across task count
  heuristic — server fallback, 1000/500, no agent input

Added token_note field to TaskUpdate schema and exposed note param on
update_task_status and record_interactive_task MCP tools.
TOOLS.md documents the full taxonomy. 185 tests pass.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-03-29 18:47:40 +02:00
tegwick
fdfd4365cd feat(token-tracking): three-tier token recording on task done 
Token events are now always created when update_task_status is called
with status="done", using the best available data:

  Tier 1 (best): exact tokens_in + tokens_out passed by agent
  Tier 2:        workplan_tokens_in + workplan_tokens_out prorated
                 across workstream task count (note="workplan")
  Tier 3 (fallback): heuristic 1000 in / 500 out (note="heuristic")

Non-done status changes never create a token event.
MCP tool updated with workplan_tokens_in/out params and tiered docs.
Ralph-workplan skill files updated with the three-tier guidance.
184 tests pass.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-03-29 18:28:18 +02:00
tegwick
58e1bafce9 feat(token-tracking): record AI token consumption per task (CUST-WP-0029) 
Introduces end-to-end token consumption tracking so agent work is
visible as a cost/effort metric alongside tasks and workplans.

- Migration o2j3k4l5m6n7: token_events table with FK indexes on
  task_id, workstream_id, repo_id, created_at
- ORM model, Pydantic schemas (TokenEventCreate, TokenEventRead with
  computed tokens_total, TokenSummary)
- Router: POST /token-events/, GET /token-events/ (7 filters),
  GET /token-events/summary/ (task|workstream|repo|commit|release scope)
- MCP tools: record_token_event, get_token_summary (formatted table)
- update_task_status enriched with optional tokens_in/tokens_out
  passthrough — one call creates status update + token event
- Dashboard token-cost.md page: by-repo bar, by-workplan table,
  by-model bar, top-10 tasks by tokens
- ralph-workplan skill updated with token reporting guidance and
  per-task heuristics for estimating counts
- Tests: test_token_events.py + test_token_passthrough.py (182 pass)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-03-29 17:46:46 +02:00