Model selector status messages (assign, clear, fallback-chain) interpolated
`roleInfo?.name ?? role`, showing the display name ("Fast"/"Thinking") instead
of the tag ("SMOL"/"SLOW"). Aligned them with the rest of the TUI
(model-browser.ts, model-hub.ts) which use `info.tag ?? info.name ?? role`.
Fixes#5585
forwardSyncProgress copied the subagent's initial pending snapshot over the job-owned running status via a wholesale Object.assign, reverting mixed-split job rows to pending. Forward only the live metric fields (resolved model, reasoning, counters, recent activity) and leave status/identity to the job body.
Fixes#5060
Forwarded async task progress metadata into job snapshots so polling rows can render the effective resolved model and reasoning selector.
Added focused renderer coverage for enabled, disabled, malformed, and bash job rows.
Fixes#5060
- Added the `xd://` virtual device protocol (`internal-urls/xd-protocol.ts`, `tools/xdev.ts`): tools declaring `loadMode: "discoverable"` are unmounted from the request tools array and driven via `read xd://` (list/docs+schema) and `write xd://<tool>` (execute), gated by the `tools.xdev` setting (default on) and inlined into the system prompt.
- Merged the `irc`, `job`, and `launch` tools into a single `hub` tool (`tools/hub/`, `async/job-manager.ts`): messaging keeps `send`/`inbox`/`list`, job control maps to `wait`/`cancel`/`jobs`, process supervision keeps `start`/`logs`/`stop`/`restart`/`describe` with `ps`, and the unified `wait` races background jobs against peer messages; SDK `IrcTool`/`JobTool`/`LaunchTool` are replaced by `HubTool`.
- Removed the hidden `resolve` tool in favor of the `xd://resolve`/`xd://reject`/`xd://propose` resolution devices, auto-including `write` whenever a deferrable tool or plan mode is present.
- Removed the BM25 tool-discovery system: the `search_tool_bm25` tool, the `tool-discovery` module, the `tools.discoveryMode`/`mcp.discoveryMode`/`mcp.discoveryDefaultServers`/`tools.essentialOverride` settings, per-tool MCP selection, and the `mcp_tool_selection` message type.
- Unified tool presentation on `ToolLoadMode` (`essential`|`discoverable`), replacing the custom-tool `xdev?: boolean` opt-out; custom, extension, MCP, RPC host, image-generation, and TTS tools now default to `discoverable`, and added a `satisfies` predicate to `SoftToolRequirement`.
- Removed the standalone `ssh` command tool and `ssh/ssh-executor` (the `ssh://` read/write/search protocol stays), and made `--tools` address hidden built-ins.
- Updated collab-web to render `xd://` dispatches and `hub` op families, dropped the `search_tool_bm25`/`ssh`/`report-finding` renderers, refreshed tool docs and prompts, and migrated the affected tests and changelogs.
- Renamed the system prompt's project-context section wrapper from `<context>` to `<repo-rules>` in `project-prompt.md`, so it no longer collides with the `task` tool's `context` parameter under in-band XML tool dialects.
- Updated `snapcompact-inline`'s context-section strip pattern to match `<repo-rules>`.
- Updated the `snapcompact-inline` test fixture to the new wrapper.
- Changelog entry for this rename rides with the fused device-migration commit (adjacent changelog lines).
- Added the `enforceSeenLines` option to hashline `PatcherOptions` (defaults `true`); the seen-line guard in `Patcher` now runs only when enabled.
- Added the `edit.enforceSeenLines` coding-agent setting (default off) and wired it through `edit/hashline/execute.ts` into the `Patcher`.
- Stopped `file-snapshot-store` excluding column-clipped (>512-char) lines from a snapshot's seen set, so single-line edits on long lines apply without a full-width re-read.
- Updated `seen-line-guard` tests and the hashline/coding-agent changelogs.
The re-entry system prompt led with "Read the existing plan" and its
"different task -> overwrite it" step contradicted the planExists guidance
("leave that plan in place and start a fresh file"). Weak models fixated on
reconciling the incomplete old plan and dropped the new request entirely.
- Rewrote the Re-entry procedure in plan-mode-active.md to treat the new
request as the primary input and the old plan as reference only, and to
fold corrections for unfinished old work INTO the new plan rather than
substituting them for the new request.
- Aligned the "different task" branch with the planExists section (fresh
file, no overwrite), removing the contradiction.
- Added reentry-prompt.test.ts asserting the anchoring contract and the
absence of the contradiction.
Fixes#5576
- Added `interruptMode: never` to all updated Go, Rust, and TypeScript built-in discovery rule files.
- Changed `ts-no-inline-cast-access.md` from `interruptMode: tool-only` to `interruptMode: never`.
- Added a `plan-mode-active.md` directive barring changelog/release notes, doc updates, formatter/linter runs, and scaffolding removal as plan steps, so plan mode stops enumerating the mechanical cleanup tail that runs automatically after the change works.
- Clarified that behavior-defining tests and the end-to-end proof stay in the **Verification** section rather than being dropped as cleanup.
- Restored the compact two-row `Editor` layout by default and gated the dedicated IME-safe bottom border behind `setImeSafeCursorLayout()`.
- Added `tui.imeSafeCursor` as an opt-in appearance setting and applied it to initial and replacement editors.
- Added regression coverage for compact default rendering while retaining terminal-local IME preedit protection.
- Updated the TUI changelog for the opt-in compatibility layout.
Generated per-module loaders instead of eagerly evaluating the entire bundled compatibility graph during extension bootstrap.
This prevents appserver startup from cycling through its own retained command modules before the Unix socket is created.
Fixes#5568
- Parsed nested marked text tokens through the inline parser.
- Added browser-template regression coverage for ordered, unordered, and nested lists.
Fixes#5567
The previous follow-up rooted path-like commands at the resolved cwd in
the shared resolvePluginStdioPaths helper, which regressed plugin
.mcp.json semantics: plugin commands are relative to the plugin package
root, so a plugin shipping ./bin/server with cwd="work" would resolve to
<pkg>/work/bin/server and ENOENT. Add a commandBase parameter defaulting
to "config-dir" (the plugin contract) and pass "cwd" only from the Codex
importer, where the OS resolves the relative command against the spawned
process cwd.
Fixes#5561
resolvePluginStdioPaths resolved a path-like command against the config
directory unconditionally, but the stdio transport spawns the subprocess
with the rooted cwd as its process cwd, so the OS resolves a relative
command from there. For cwd="server", command="./bin/mcp" that meant OMP
looked for <configDir>/bin/mcp instead of <configDir>/server/bin/mcp.
Root cwd first, then resolve path-like commands from that rooted cwd,
falling back to configDir only when no cwd is set.
Fixes#5561
The Codex config.toml importer copied only command/args/url into the
returned MCPServer, dropping cwd and leaving relative command values
verbatim. MCP stdio spawning resolved those against the session cwd, so
the bundled Codex Computer Use server (relative command, cwd = ".") failed
with ENOENT. Route command/cwd through resolvePluginStdioPaths against the
config directory, matching the claude-plugins/omp-plugins fix in #5481.
Fixes#5561
Each tool-result turn now re-arms one text-only continuation while prewalk is pending. Consecutive prose replies without intervening tool progress terminate naturally, preserving multi-step plan detours without restoring the completion loop.
Fixes#5551
A prose plan and a bash-only completion are structurally identical after any pre-implementation tool detour, so the net now stays armed across every such turn and fires exactly once on the first text-only reply. This bridges read/record/bash detours before the plan while bounding a genuine no-edit completion to a single continuation instead of looping.
Fixes#5551
The continuation net now stays armed across a todo-only turn and disarms only when a non-planning tool runs without a prose plan, so the normal plan-nudge to todo to prose to edit flow still reaches implementation while a bash-only completion no longer loops.
Fixes#5551
Used centralized internal-URL detection so filesystem-only streaming edit work does not resolve ssh:// paths through the cwd.
Added a replace-mode regression covering controlled tool dispatch for an ssh:// path.
Fixes#5552
Limited the hidden continuation safety net to the assistant turn immediately following the plan nudge, so later bash-only completion ends normally.
Added regression coverage for commit-style flows that never call edit or write.
Fixes#5551
The /omfg generated-rule path validated conditions with raw new RegExp in normalizeConditionRegex and isValidRegexCondition, so a generated rule with a leading (?i)/(?m)/(?s) inline flag was rejected as "Invalid condition regex" before reaching TtsrManager.addRule. Route all three validation sites through compileRuleCondition.
Fixes#4796
- Updated prewalk gating in `AgentSession` to key the todo gate on active tools instead of registry presence, so deactivated todo tools no longer block prewalk handoff.
- Changed subprocess tool filtering so `todo` is stripped for normal subagents but retained when prewalk is armed, and propagated the prewalk state through tool-session setup.
- Added regression tests for restricted active-tool slates and prewalk/non-prewalk subagent tool propagation to verify todo is handled correctly in each case.
- Updated system and todo prompt templates to require batching todo tool calls with real action calls instead of sending them alone.
- Passed a new `prewalkArmed` session flag in `createAgentSession`, set from whether `prewalk` was supplied.
Bash command timeouts now render with a warning (yellow) border instead
of an error (red) border, reflecting that the timeout ran its course
rather than the command failing.
The timeout is no longer thrown as a ToolError — instead #buildCompletedResult
returns a non-throwing error result (isError=true, keeping the model-facing
contract) with details.timedOut=true. The renderer reads this flag to pick
state="warning" (yellow) instead of state="error" (red).
The timedOut flag is propagated from bash-executor.ts, which now sets
timedOut=true on timeout return paths and leaves it unset on user-abort
paths. This distinguishes timeouts from user Esc-cancels — previously both
returned cancelled=true with no way to tell them apart in bash.ts.
- Simplified `bash` guidance to tighten allowed command patterns, pipeline limits, and launch-based process handling.
- Reworked `browser` instructions into grouped helper sections while preserving selector restrictions and key action semantics.
- Harmonized `eval`, `irc`, `read`, and `todo` prompt wording around state reuse, messaging, selector formats, and task operations.
robomp-bot