Commit Graph

19 Commits

Author SHA1 Message Date
TechDufus 890dd88597 fix(coding-agent): make mixed-agent task batches atomic and inspectable 2026-07-23 16:47:51 -05:00
can1357 49782ecce6 Merge PR #6326: feat(coding-agent): configure isolated task apply behavior (@korri123) 2026-07-23 11:48:54 +02:00
panosAthDBX bfef3f7eea fix(task): reject sparse model fallback arrays 2026-07-23 09:53:54 +01:00
panosAthDBX 10e1ba911c test(task): cover model override precedence 2026-07-23 09:50:25 +01:00
panosAthDBX 1c8d9db6dd feat(task): allow per-call model selection 2026-07-23 09:42:39 +01:00
Kormákur c2fde74ef0 feat(coding-agent): configure isolated task apply behavior 2026-07-23 00:48:48 +00:00
can1357 242bcbef8c revert #6130: doesn't really work well 2026-07-22 23:54:15 +02:00
can1357 9c6c53435b Merge PR #6130: feat(coding-agent): add capture-only apply control to the task tool (@korri123) 2026-07-22 21:13:23 +02:00
Kormákur 1b83629ef1 fix(coding-agent): enforce capture-only isolation 2026-07-22 12:28:46 +00:00
roboomp 05668faa5f fix(task): surfaced actionable shape error for batch calls
The flat single-spawn task wire schema carries arktype `"+": "delete"`, so a
batch `{ context, tasks[] }` payload sent while `task.batch` is disabled has
those keys stripped and is then rejected as `task must be a string (was
missing)` in the agent loop. That preempts the tool's own actionable checks
(validateShapeParams / validateSpawnParams), so the model only ever saw the
misleading arktype error instead of "task.batch is disabled…".

Mark TaskTool with lenientArgValidation so the agent loop forwards the raw
args to execute() on any arktype failure, letting the tool's shape checks
surface the real reason. Valid calls still normalize through arktype; the
success path is unchanged. Mirrors the existing yield-tool pattern.

Fixes #6039
2026-07-21 20:41:14 +00:00
Kormákur 6f75cf0d5d feat(coding-agent): add task tool capture-only apply control
Add an optional `apply` parameter to the `task` tool so
`isolated: true, apply: false` captures patch/branch artifacts without
applying changes to the parent checkout. Available as a flat top-level
control and per `tasks[]` item. Shares the task/eval isolation-to-executor
translation via a single `toStructuredSubagentIsolationControls` adapter.
2026-07-20 18:18:23 +00:00
roboomp 83bfb6dbd8 fix(task): avoided boolean output subschemas
Represent task output schema inputs as explicit JSON-compatible types so ArkType emits object-form subschemas accepted by llama.cpp grammar generation.

Fixes #5957
2026-07-18 04:11:26 +00:00
vmcall d944879f21 feat(task): unified structured subagent execution
- Added per-invocation task schemas with strict and permissive validation.
- Shared task and eval agent policy, artifacts, isolation, and lifecycle handling.
- Enabled host-restricted plan-mode eval agents and persisted their capability clamp.

Fixes #5279
2026-07-17 17:36:59 +02:00
can1357 cb2153e9a5 feat(coding-agent-task): implemented agent-centric flat task structure
- Renamed task wire fields, replacing `assignment` and `description` with `task` and `name` while removing `role` references.
- Implemented automated task UI label generation using a tiny model to replace manual role descriptions.
- Updated task execution and rendering logic to support per-item agent resolution and dynamic badge display.
- Migrated schemas, prompts, and test suites to enforce the new flat task structure and agent-centric policy.
2026-07-11 13:44:04 +02:00
roboomp 8e2dfcc43e fix(agent): routed acquire-time abort through queued-spawn settled path
When a batched task spawn is cancelled while still queued behind task.maxConcurrency the semaphore now rejects acquire(), but the previous patch let the abort throw past the aborted handler so progress.status and onSettled never fired and buildAsyncDetails kept reporting the batch as running. The wrapper now records whether the slot was held, funnels both acquire-time and post-acquire aborts through the same aborted branch (releasing only when held), and a batch regression test pins the contract.

Fixes #3930
2026-06-30 23:46:19 +00:00
can1357 d46bd0139b refactor(coding-agent): fixed system prompt customization path
- Fixed `SYSTEM.md` integration to correctly include custom-rendered sections like rules and skills.
- Consolidated system prompt validation by requiring `<skills>` tag presence instead of specific prose.
- Removed redundant system prompt math-formatting tests and orphaned task batch documentation tests.
2026-06-19 16:16:37 +02:00
can1357 1fa4f2eee8 fix(subagent-session-spawning): resolved subagent parent id propagation
- Forwarded `parentAgentId` through task and eval launch paths when spawning subagents.
- Mapped `parentAgentId` to `parentId` in `createAgentSession`.
- Passed each caller's session `getAgentId` (or `MAIN_AGENT_ID`) as the parent for spawned agents.
2026-06-16 23:03:54 +02:00
can1357 844c8dbdfe fix(coding-agent): restored task sync mode when async.enabled was false
- Updated TaskTool to skip `session.asyncJobManager` and run `task` spawns inline whenever `async.enabled` is false.
- Set `async.enabled` default to `true` and updated task prompts/settings text to reflect async-versus-sync behavior.
- Adjusted task batching tests to cover both async background execution and synchronous batched execution when async is disabled.
2026-06-11 16:07:38 +02:00
can1357 72c12ff9c2 feat(task): migrated task tool to batch-first mode with shared context
- Replaced task-simple-mode with a `task.batch` setting enabled by default.
- Updated task schema to use batch `{agent, context, tasks[]}` payloads.
- Migrated task execution to spawn one async job per task and merge outputs.
- Removed per-call schema passing while preserving legacy flat task calls.
2026-06-10 23:47:28 +02:00