machina-cli 0.2.27__tar.gz → 0.4.2__tar.gz

{machina_cli-0.2.27 → machina_cli-0.4.2}/CHANGELOG.md

@@ -2,6 +2,33 @@
 
 All notable changes to machina-cli are documented here.
 
+## [0.4.2] - 2026-06-29
+
+### Added
+- **`machina org usage`**: roll up LLM token consumption across an organization's agent executions — the repeatable answer to "how many tokens did this org consume". Token usage is recorded on **agent** executions (`execution_tokens`), not workflow executions, and the Client-API `execution/agent-search` totals are **page-level only**, so the command iterates the org's projects and paginates+sums client-side, broken down by **project, agent, and day**, with the prompt/completion split surfaced (the cost-shape signal — SportingBOT's chat is ~98% prompt tokens). Flags: `--org` (defaults to the selected org), `--project` to scope to one, `--days` window (default 30; a frozen upper bound keeps pagination stable as new runs arrive), `--top`, `--limit`, and `--json`. Unreachable/undeployed projects are skipped and listed under `projects_unreachable` rather than failing the run. Verified end-to-end against the live Entain SportingBOT deployment.
+
+## [0.4.1] - 2026-06-26
+
+### Added
+- **Loop retry-with-critique (Cap 8.2).** When the gate passes but the independent evaluator *rejects* a turn's answer, the loop now does ONE bounded repair pass — feeding the rejection reason back into a `loop-repair` prompt — then re-verifies with `loop-evaluate-2` before deciding `idle` vs `needs_review`. Generator/evaluator → generator/evaluator/**repairer**. A *gate* failure is still never repaired (straight to `needs_review`); the repair is bounded to one pass (no loop). `value.verification.repaired` records whether a turn self-healed; the CLI appends `· self-repaired` to the verdict line. Provisioner adds the `loop-repair` + `loop-evaluate-2` prompts.
+
+## [0.4.0] - 2026-06-26
+
+### Added
+- **Loop verification — the generator/evaluator separation (Cap 8).** A loop turn is now finalized `idle` only after it clears a **deterministic gate** (cheap, code-only — a non-trivial answer, no error marker, and the tool succeeded if one ran) **and** an **independent evaluator** (`loop-evaluate`: a separate prompt with a fresh context + an "assume it's wrong" posture and its own `EVAL_MODEL`). Any failure → **`needs_review`**, the human checkpoint — never a silent pass. The resume path carries an attempt budget (`LOOP_MAX_ATTEMPTS`) so the beat can't re-run a stuck session forever. Verified live against a staging pod.
+  - `machina loop` (`run`/`watch`/`say`) now treats `needs_review` as terminal and renders the evaluator's verdict (`✓ verified` / `⚠ needs review — <reason>`).
+  - Provisioner `docs/harness-loop-kit/provision.py` gains the `loop-evaluate` prompt, the gate, and the attempt budget; tunable via `EVAL_MODEL` (use a stronger model than the generator in prod) and `LOOP_MAX_ATTEMPTS`.
+  - New: `docs/harness-loop-kit/PLAYBOOK-SCORECARD.md` — the loop scored against the Loop-Engineering playbook (the five failure modes, the First-Loop Checklist, the Stripe-Minions pattern) with the live verification results.
+
+## [0.3.0] - 2026-06-26
+
+### Added
+- **`machina loop …`**: drive a durable agentic turn loop (harness) that runs server-side on Studio primitives, with the CLI as a thin driver/observer (same pattern as `machina factory`). Commands: `run`, `watch`, `say`, `stop`, `sessions`.
+  - **Durable**: the loop is a scheduled agent + `harness_session` documents; each turn is persisted before advancing, and the beat resumes a session left `active` (survives crash / async tool / awaiting input). A turn ends at `idle` (awaiting the next `say`).
+  - **Multi-turn**: `say` injects a follow-up; prior conversation is fed back to the model. `watch` is turn-aware and streams new turns (incl. `→ tool` / `← result` steps).
+  - Server-side resources (prompt/workflow/agent) live in the project's Studio; the CLI reuses the existing project session — no new credential.
+  - Architecture + chapter-by-chapter build (caps 1–7) and verified platform contracts: `docs/agentic-harness-loop.md`. Dynamic multi-tool dispatcher reference: `docs/loop-tools-connector.py`.
+
 ## [0.2.27] - 2026-06-25
 
 ### Added

{machina_cli-0.2.27 → machina_cli-0.4.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: machina-cli
-Version: 0.2.27
+Version: 0.4.2
 Summary: CLI for the Machina AI Agent platform
 Project-URL: Homepage, https://machina.gg
 Project-URL: Repository, https://github.com/machina-sports/machina-cli
@@ -144,8 +144,16 @@ machina org list --page 2              # Page 2
 machina org list --json                # Output as JSON
 machina org create <name>              # Create organization
 machina org use <org-id>               # Set default organization
+machina org usage                      # Token consumption (last 30d), by project/agent/day
+machina org usage --days 7             # Narrow the window
+machina org usage --project <id>       # Scope to one project (faster)
+machina org usage --json               # Output as JSON
 ```
 
+`org usage` rolls up LLM token consumption from agent executions (`execution_tokens`)
+across the organization's projects — broken down by project, agent, and day, with the
+prompt/completion split. Useful for cost/capacity tracking.
+
 ### Projects
 
 ```bash

{machina_cli-0.2.27 → machina_cli-0.4.2}/README.md

@@ -112,8 +112,16 @@ machina org list --page 2              # Page 2
 machina org list --json                # Output as JSON
 machina org create <name>              # Create organization
 machina org use <org-id>               # Set default organization
+machina org usage                      # Token consumption (last 30d), by project/agent/day
+machina org usage --days 7             # Narrow the window
+machina org usage --project <id>       # Scope to one project (faster)
+machina org usage --json               # Output as JSON
 ```
 
+`org usage` rolls up LLM token consumption from agent executions (`execution_tokens`)
+across the organization's projects — broken down by project, agent, and day, with the
+prompt/completion split. Useful for cost/capacity tracking.
+
 ### Projects
 
 ```bash