Commit Graph

2773 Commits

Author SHA1 Message Date
Hunter B d47431269f feat(fleet): FleetExecutor runs workers as real exec subprocesses (#3096/#3154)
Adds FleetExecutor: drives a worker as a local `codewhale exec` subprocess via
the existing host adapter, incrementally drains its stream-json output into
FleetWorkerEventPayload ledger events, and finalizes the terminal outcome from
the process exit. The worker's heavy runtime/tool construction lives in its own
process, so the orchestrator only ingests a compact event stream — the
isolation pattern that keeps fanout light (per Codex/Kimi/Claude Code).

Verified end-to-end by an integration test that runs a REAL subprocess emitting
stream-json (standing in for `codewhale exec`) through the real adapter and
asserts RunningTool + terminal Completed events flow out — no codewhale binary
needed. 8 executor tests pass; the 58 existing fleet tests stay green
(executor is not yet wired into `codewhale fleet run`, so no behavior change).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-13 01:24:22 -07:00
Hunter B 91cbaf2811 feat(fleet): exec-subprocess worker building blocks (#3096/#3154)
A fleet worker IS a headless `codewhale exec` run — not a separate execution
engine. New crates/tui/src/fleet/executor.rs provides the bridge:

- build_worker_exec_command: FleetTaskSpec + FleetExecConfig ->
  `codewhale exec --auto --output-format stream-json [--model/--allowed-tools/
  --disallowed-tools/--max-turns/--append-system-prompt] <prompt>`. Secrets are
  never on argv (worker resolves its own config/keyring).
- map_exec_stream_line: maps the worker's stream-json events
  ({"type":"tool_use"|"content"|"tool_result"|"done"|"error"}) into
  FleetWorkerEventPayload so the durable ledger persists the worker's OWN event
  vocabulary (collapses the translation layer toward identity).
- classify_worker_exit: process exit -> terminal Completed/Failed/Cancelled.

This is the substrate for replacing the local simulation with real headless
workers. fleet_task_prompt is now pub(crate). 7 new unit tests; building blocks
only (no behavior change yet) so the 58 existing fleet tests stay green.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-13 01:20:28 -07:00
Hunter B b1cc6ecf5d merge: fleet security/trust + headless-worker foundation + depth unification 2026-06-13 01:11:50 -07:00
Hunter B e8b52ac57a feat(fleet): security/trust + headless-worker foundation; unify recursion depth
Lands the Agent Fleet security/trust boundary and the headless-worker bridge on
the v0.8.60 line, and collapses the sub-agent and fleet recursion model into a
single shared axis (Hunter steer: "not two moving targets").

Security & trust (#3165):
- FleetTrustLevel, FleetSecurityPolicy, FleetSecretRef (redacted), FleetWorkerAuth,
  FleetCapabilityGrant, FleetAlertEndpoint (redacted) in protocol.
- secrets: resolve_direct(key, source_hint) — fleet secret resolution, never logged.
- Host adapters refuse secret-bearing env keys; SSH uses SendEnv (no argv secrets).

Roles & delegation (#3167):
- fleet role -> SubAgentType mapping; reviewer/verifier default read-only.

Headless worker bridge (#3096/#3154, partial — still simulation, real spawn next):
- worker_runtime: FleetTaskSpec -> AgentWorkerSpec, status -> ledger events,
  exec hardening (mirrors #3027), parallel-safe read-only tool set (#2983).
- FleetManager carries an optional SharedSubAgentManager + exec config.

Recursion depth — ONE axis:
- codewhale_config now owns DEFAULT_SPAWN_DEPTH (3) + MAX_SPAWN_DEPTH_CEILING (3).
- sub-agent DEFAULT_MAX_SPAWN_DEPTH and the fleet clamp both source these consts.
- fleet default raised 1 -> 3 to match standalone sub-agents; root runs at depth 0,
  budget gates child delegation. End-to-end test proves a depth-0 fleet worker
  reaches 3 nested levels (afford >= 3).

Dogfood scaffolding (#3166, partial): docs/examples/fleet-dogfood.toml.

Tests green: codewhale-config fleet, codewhale-tui fleet (58), subagent max_depth;
cargo fmt + git diff --check clean; cargo check --workspace ok.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-13 01:10:30 -07:00
Hunter B 7c52d6ae11 Merge provider model refresh for v0.8.60 2026-06-12 23:57:54 -07:00
Hunter B 754cff1c84 Refresh GLM and MiniMax provider models 2026-06-12 23:56:01 -07:00
Hunter B 344e89c93d merge #3082 Checklist strategy workflow 2026-06-12 22:55:23 -07:00
Hunter B 751ff943aa fix(tui): group strategy context under work checklist 2026-06-12 22:55:11 -07:00
Hunter B 467bb309dd merge #3142 Agent run receipts 2026-06-12 22:46:08 -07:00
Hunter B c0ba6ce5ad feat(subagents): add agent run receipts 2026-06-12 22:45:53 -07:00
Hunter B 07670871d0 merge #3164 Fleet manager runbook skill 2026-06-12 22:26:49 -07:00
Hunter B fb1a27b4f2 feat(fleet): add manager runbook skill 2026-06-12 22:26:35 -07:00
Hunter B 33f112143d merge #3163 Runtime SDK fleet helpers 2026-06-12 22:17:40 -07:00
Hunter B 26925ae644 feat(runtime-sdk): add fleet helper client
Refs #3163.

Adds the @codewhale/runtime-sdk workspace with typed fleet Runtime API helpers, protocol-shaped TypeScript declarations, JSON/SSE event fixture handling, and typed RuntimeCapabilityError failures for create/event-stream endpoints that the Rust API has not exposed yet.

Documents the SDK contract in docs/RUNTIME_API.md and wires npm workspace verification through npm test --workspace @codewhale/runtime-sdk.
2026-06-12 22:17:16 -07:00
Hunter B e8f6816472 merge #1310 MiniMax provider route 2026-06-12 22:05:52 -07:00
Hunter B e9f0c1c13b feat(config): add first-party MiniMax provider route
Refs #1310.

Adds the direct minimax provider slot, auth/config/env bindings, model aliases, docs, and picker/status surfaces. MiniMax requests now set reasoning_split and preserve reasoning_details history so thinking stays out of answer text.

Also updates Moonshot/Kimi reasoning classification for the Kimi Code route so reasoning_content streams into Thinking cells instead of inline prose.
2026-06-12 22:05:22 -07:00
Hunter Bown e6005eb9ac feat(config): add first-party Z.ai and StepFlash/StepFun provider routes (#3187) (#3191)
* feat(config): add first-party Z.ai and StepFlash/StepFun provider routes (#3187)

Add Z.ai (GLM Coding Plan) and StepFun/StepFlash as first-party providers:

Provider: Z.ai
- Default model: GLM-5.1 (200K context, 128K output, thinking, function calling)
- Base URL: https://api.z.ai/api/coding/paas/v4
- Auth: ZAI_API_KEY / Z_AI_API_KEY
- Config key: [providers.zai]

Provider: StepFun
- Default model: step-3.7-flash (256K context, 256K output, 3-level reasoning)
- Base URL: https://api.stepfun.ai/v1
- Auth: STEPFUN_API_KEY / STEP_API_KEY
- Config key: [providers.stepfun]

Both providers use OpenAI-compatible Chat Completions transport.

Includes:
- ProviderKind enum variants with serde aliases
- ProvidersToml config fields with aliases
- Provider trait implementations in provider.rs
- PROVIDER_REGISTRY, ALL, KIND_LOOKUP, FROM_KIND_LOOKUP updates
- EnvRuntimeOverrides fields and env-var loading
- TUI ApiProvider enum and lookup table updates
- CLI provider_slot and provider_env_vars additions
- Exhaustive match arms across client.rs, config.rs, engine.rs, main.rs,
  provider_picker.rs, ui.rs, and config_persistence.rs

Verified: cargo check passes, config provider tests (67/67) pass.

* test(tui): update provider picker tests for Zai and Stepfun additions
2026-06-12 19:58:57 -07:00
Hunter B 12d3053a15 merge #3162 fleet status surfaces 2026-06-12 19:35:31 -07:00
Hunter B 6b5ffd5627 feat(tui): expose fleet status surfaces 2026-06-12 19:35:11 -07:00
Hunter B 86c87ca6a4 merge #3161 fleet alerts 2026-06-12 19:18:01 -07:00
Hunter B bc61857fad feat(tui): add opt-in fleet alert dry-runs 2026-06-12 19:17:50 -07:00
Hunter B a656175c49 merge #3160 fleet task specs 2026-06-12 19:01:14 -07:00
Hunter B 714276e1c1 feat(tui): add verifiable fleet task specs 2026-06-12 19:01:00 -07:00
Hunter B 8e53aac3b5 merge #3159 fleet scheduler policy 2026-06-12 18:43:44 -07:00
Hunter B 44f429f108 feat(tui): add fleet scheduler policy 2026-06-12 18:43:36 -07:00
Hunter B c78d3b63fc merge #3158 fleet host adapters 2026-06-12 18:31:39 -07:00
Hunter B 9e5688b8c8 feat(tui): add fleet host adapters 2026-06-12 18:31:29 -07:00
Hunter B e35684efed merge #3157 fleet manager commands 2026-06-12 18:15:26 -07:00
Hunter B 66a45c0070 feat(tui): add local fleet manager commands 2026-06-12 18:15:18 -07:00
Hunter B 8ba8f7f2f5 merge #3156 fleet ledger 2026-06-12 17:56:41 -07:00
Hunter B 023434cb2a fix(tui): harden fleet ledger replay and compaction 2026-06-12 17:56:26 -07:00
Hunter B b0c876e5ac merge #3155 fleet protocol fixes into ledger branch
# Conflicts:
#	crates/protocol/src/fleet.rs
2026-06-12 17:50:45 -07:00
Hunter B 59848f18fc merge #3155 fleet protocol types
# Conflicts:
#	crates/protocol/src/lib.rs
2026-06-12 17:50:18 -07:00
Hunter B f0ae838eb1 fix(protocol): harden fleet JSON contract 2026-06-12 17:48:59 -07:00
Hunter B 5b47d46092 chore: seed v0.8.60 branch from main after v0.8.59
# Conflicts:
#	README.md
#	crates/cli/src/lib.rs
#	crates/config/src/lib.rs
#	crates/config/src/provider.rs
#	crates/tui/src/commands/mod.rs
#	crates/tui/src/localization.rs
#	crates/tui/src/prompts.rs
#	crates/tui/src/tui/ui.rs
#	crates/tui/src/tui/ui/tests.rs
#	crates/tui/src/tui/widgets/tool_card.rs
#	scripts/benchmarks/cli-compare.py
2026-06-12 17:42:22 -07:00
CodeWhale Agent 3d822488fe chore(fmt): apply cargo fmt for CI parity gate 2026-06-12 17:02:17 -07:00
CodeWhale Agent 97f8a98b75 fix(config): align Kimi OAuth credentials with Kimi Code 2026-06-12 16:43:25 -07:00
CodeWhale Agent 738a265197 fix(tui): show Bash for shell work in visible UI 2026-06-12 16:32:32 -07:00
CodeWhale Agent f870eb3e19 fix(tui): keep subagents alive across backgrounded waits 2026-06-12 16:16:39 -07:00
CodeWhale Agent d48373617f fix(tui): clear sidebar hover highlight on exit while loading 2026-06-12 15:54:08 -07:00
CodeWhale Agent b0577057c9 fix(tui): localize notifications and persist sidebar focus 2026-06-12 15:47:44 -07:00
dependabot[bot] 7856d26d64 build(deps): bump tempfile from 3.24.0 to 3.27.0 (#2997)
Bumps [tempfile](https://github.com/Stebalien/tempfile) from 3.24.0 to 3.27.0.
- [Changelog](https://github.com/Stebalien/tempfile/blob/master/CHANGELOG.md)
- [Commits](https://github.com/Stebalien/tempfile/compare/v3.24.0...v3.27.0)

---
updated-dependencies:
- dependency-name: tempfile
  dependency-version: 3.27.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-06-12 15:44:26 -07:00
dependabot[bot] d818ac4e2a build(deps): bump image from 0.25.9 to 0.25.10 (#3000)
Bumps [image](https://github.com/image-rs/image) from 0.25.9 to 0.25.10.
- [Changelog](https://github.com/image-rs/image/blob/v0.25.10/CHANGES.md)
- [Commits](https://github.com/image-rs/image/compare/v0.25.9...v0.25.10)

---
updated-dependencies:
- dependency-name: image
  dependency-version: 0.25.10
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-06-12 15:44:19 -07:00
dependabot[bot] 45f334828f chore(deps): bump pdf-extract from 0.7.12 to 0.10.0 (#3184)
Bumps [pdf-extract](https://github.com/jrmuizel/pdf-extract) from 0.7.12 to 0.10.0.
- [Commits](https://github.com/jrmuizel/pdf-extract/compare/v0.7.12...v0.10.0)

---
updated-dependencies:
- dependency-name: pdf-extract
  dependency-version: 0.10.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-06-12 15:44:13 -07:00
dependabot[bot] 9373c5d6b2 chore(deps): bump flate2 from 1.1.5 to 1.1.9 (#3185)
Bumps [flate2](https://github.com/rust-lang/flate2-rs) from 1.1.5 to 1.1.9.
- [Release notes](https://github.com/rust-lang/flate2-rs/releases)
- [Commits](https://github.com/rust-lang/flate2-rs/compare/1.1.5...1.1.9)

---
updated-dependencies:
- dependency-name: flate2
  dependency-version: 1.1.9
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-06-12 15:44:06 -07:00
dependabot[bot] ec5b0791be chore(deps): bump objc2 from 0.6.3 to 0.6.4 (#3186)
Bumps [objc2](https://github.com/madsmtm/objc2) from 0.6.3 to 0.6.4.
- [Commits](https://github.com/madsmtm/objc2/compare/objc2-0.6.3...objc2-0.6.4)

---
updated-dependencies:
- dependency-name: objc2
  dependency-version: 0.6.4
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-06-12 15:43:59 -07:00
CodeWhale Agent e4989c0eae feat(tui): add Copy to the sidebar right-click context menu
Removing Paste from the sidebar menu (#3065) left rows with no copy
path at all — sidebar text can't be mouse-selected. Right-clicking a
sidebar row now offers Copy, which writes the row's untruncated text
plus its hover detail to the clipboard via a new
ContextMenuAction::CopyText. Run stays first for clickable rows.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-12 15:35:20 -07:00
CodeWhale Agent ba1104251a feat(tui): richer Work overflow and Agents hover detail in sidebar
- Work panel: hovering the '+N more checklist items' overflow row now
  reveals the omitted checklist items in the popover, so a
  height-constrained panel no longer hides work items with no way to
  inspect them (#3063).
- Agents panel: the compact agent label row's hover text now carries a
  full dossier — id, role, status, elapsed time, step count, objective
  (new SidebarAgentRow field, from SubAgentAssignment), branch, and
  untruncated progress. Compact rows stay unchanged.
- Detail-row hover keeps the full progress text instead of the
  summarized form.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-12 15:26:53 -07:00
CodeWhale Agent 2da01367e6 fix(tui): keep sidebar resize live during active turns
The loading mouse filter (should_drop_loading_mouse_motion) dropped all
Drag events while app.is_loading unless a transcript selection or
scrollbar drag was active. A sidebar resize started on the handle
(Down passes the filter) then never received its Drag events, leaving
the resize wedged mid-gesture during live runs (#3063, symptom of the
#3096 subagent-runtime pressure on the TUI).

- Allow Drag events through the loading filter while
  app.sidebar_resizing is set.
- Clear last_sidebar_area / last_sidebar_handle_area and any in-flight
  resize when the sidebar is hidden or doesn't fit, so stale handle
  hit-areas can't capture clicks.
- Tests: resize down/drag/up while loading, mouse-up outside the
  handle still ends the resize.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-12 15:26:53 -07:00
CodeWhale Agent 3169c459ed docs(changelog): document harvested v0.8.59 features 2026-06-12 14:54:47 -07:00