Wired the generic discovery methods on AgentSession into the tool
factory session, resolved the effective discovery mode (tools.discovery
Mode wins; mcp.discoveryMode as back-compat alias for "mcp-only"),
and threaded the resulting flag into rebuildSystemPrompt so the prompt
template fires the discovery hint for both legacy and new modes.
Added the load-bearing filter in createAgentSession: when the
effective mode is "all", drop any built-in tool whose loadMode is
"discoverable" from the initial tool set unless it is essential per
computeEssentialBuiltinNames, was explicitly listed via
options.toolNames, or was restored from persistence. The model finds
hidden tools via search_tool_bm25 and activates them on demand.
Built-in activation persistence is intentionally limited to the
existing MCP persistence store for this PR; full discovered-tool
persistence is a follow-up.
The internal class is now SearchToolsTool but the wire name stays
"search_tool_bm25" so persisted MCP selections in user session files
continue to resolve correctly.
The BM25 corpus is now drawn from the generic DiscoverableTool list
(built-ins, MCP, extension, custom) rather than only MCP tools.
createIf fires when either tools.discoveryMode !== "off" or the
legacy mcp.discoveryMode is true.
AgentSession gains generic discovery methods (isToolDiscoveryEnabled,
getDiscoverableTools, getDiscoverableToolSearchIndex,
getSelectedDiscoveredToolNames, activateDiscoveredTools). The existing
MCP-named methods are kept as thin shims that filter by
source === "mcp" so existing callers continue to work.
Updated the tool's prompt copy to advertise discovery across all
sources rather than MCP only. Tests extended for the new shape.
Added two settings:
- tools.discoveryMode ("off" | "mcp-only" | "all", default "off")
- tools.essentialOverride (string[], default empty)
Converted BUILTIN_TOOLS from Record<string, ToolFactory> to
Record<string, BuiltinEntry> with a per-tool loadMode ("essential"
or "discoverable") and an optional summary used as the BM25 corpus
entry when discovery hides the tool.
Marked read, bash, edit as essential. Marked the remaining 24 built-in
tools as discoverable with hand-written summaries. search_tool_bm25
stays essential (always loaded when discovery is on; gated separately
by isToolAllowed).
Added DEFAULT_ESSENTIAL_TOOL_NAMES, computeEssentialBuiltinNames
(reads tools.essentialOverride with a default fallback), and
getBuiltinDiscoverableEntries (used by the search tool to build the
BM25 corpus).
mcp.discoveryMode is preserved as a back-compat alias for "mcp-only".
Moved the BM25 discovery index out of mcp/discoverable-tool-metadata.ts
into tool-discovery/tool-index.ts and generalized it to work over any
AgentTool source (built-in, MCP, extension, custom) instead of only MCP.
Added DiscoverableTool with explicit source and summary fields. Kept
the MCP-typed helpers (collectDiscoverableMCPTools, searchDiscoverable
MCPTools, etc.) as thin re-exports from the original path so existing
callers continue to work unchanged.
Tests cover the generic index, source filtering, summary fallback, and
back-compat with the legacy MCP-only API.
- Extended boundary duplicate absorption to single structural closing lines (`}`, `)`, `]`) for replacement groups.
- Added delimiter-balance validation so a boundary line drops only when the kept payload matches expected counts.
- Enabled default single-line structural boundary absorption for pure inserts while keeping multi-line absorbs gated by `autoDropPureInsertDuplicates`.
- Added a new `read.summarize.prose` setting to control markdown and plain-text read summaries.
- Updated the read tool to skip summarization for prose files unless that setting is enabled.
- Added coverage for default and enabled markdown read-summary behavior.
- Added configurable `ScanDetail` modes across native fd/glob/ast flows and exposed `size` on `GlobMatch`/types.
- Added parallel grep processing with optional workers, buffered entry visits, and direct small-file reads.
- Changed `filesWithMatches` to stop at first match per file and report `totalMatches` as matching-file count.
- Fixed grep count/offset limits, timeout checks, and cancel handling by enforcing offsets in aggregate results.
- Updated PI_GREP_WORKERS and auto-absorb/read-summarize behavior/docs, including `read.summarize.prose` and pure-insert defaults.
- Added tests for filesWithMatches counts, cancel cases, and count-mode behavior, plus updated grep benchmark imports.
- Exported DEFAULT_MATCH_LIMIT from the search tool module.
- Updated the search tool tests to import DEFAULT_MATCH_LIMIT from the module.
- Updated default-limit assertions and messages to use DEFAULT_MATCH_LIMIT instead of hardcoded values.
- Added `pr_create` to the GitHub tool schema and execution dispatch.
- Implemented `executePrCreate()` with title/fill validation, arg assembly, and temp-file body handling.
- Added optional PR result enrichment by parsing created PR URLs and formatting `gh pr view` summary fields.
- Updated `github.md` docs, changelog notes, and tests for `pr_create` mappings and validation.
- Converted systemPrompt APIs and state types to ordered `string[]` across agent, AI, and coding-agent surfaces.
- Added `normalizeSystemPrompts` and applied it to context normalization before building provider request payloads.
- Updated AI providers to emit separate normalized prompt blocks/messages instead of a single merged system prompt.
- Removed dedicated `projectPrompt` state and remapped that context into system-context buckets in session, dump, and token accounting.
- Aligned tests and changelogs to pass and assert `systemPrompt` as arrays with ordered prompt semantics.
- Added a new `edit.hashlineAutoDropPureInsertDuplicates` boolean setting with default `false` for hashline edits.
- Threaded the setting through tool execution contexts so hashline previews and execution honor the configured option.
- Changed pure-insert duplicate boundary absorption to run only when enabled and added tests for default-disabled and enabled behavior.
- Refactored directory tree child discovery into a dedicated direct-name helper.
- Used fs.readdir for non-gitignore directories instead of globbing.
- Kept child-name filtering for excluded and hidden entries in the same selection flow.
- Added DirectoryTree, DirectoryTreeOptions, and buildDirectoryTree exports for configurable tree rendering.
- Changed read tool directory output to use buildDirectoryTree with depth and exclusion limits.
- Added read.summarize settings and summary-mode parseable read output behavior when no selector is used.
- Added tests for truncated root/child listings and hidden or excluded entry filtering.
- Added `WorkspaceTree` and `buildWorkspaceTree` APIs for working-directory tree rendering with limits.
- Extended `buildSystemPrompt` and `createAgentSession` to resolve and pass workspace tree context for system prompts.
- Updated the system prompt template to include a `<workspace-tree>` section with truncation notices before append output.
- Added workspace-tree and system-prompt tests covering sorting, truncation, exclusions, and prompt ordering.
- `is_elidable_kind` now treats more AST kinds as elidable for TypeScript, Rust, and Java.
- Added summary tests for interface and class bodies in TypeScript, Rust, and Java.
- Merged brace-boundary read summary spans into a single `..` line-range representation.
- Grouped adjacent groupable nodes and emitted elision spans only after grouped runs met thresholds.
- Normalized end-line rows at column 0 and extended import-like groupable-kind rules across languages.
- Updated read-summary tests and added changelog notes for merged brace-pair `{ .. }` output behavior.
- Added a structural hash path that returns an ordinal-suffix bigram for lines containing only whitespace and braces.
- Kept line-number variation for other non-content lines while leaving content-bearing line hashes independent of index for stability across small shifts.
- Extended hashline pure-insert absorption to auto-drop 2+ duplicated boundary lines at ANCHOR, BOF, and EOF inserts.
- Surfaced a warning when pure-insert boundary lines are auto-dropped.
- Registered a new `read` command in the CLI command registry so `omp read` can be invoked.
- Implemented `runReadCommand` to execute the read tool, wrap it with meta notices, and print text or image-result blocks.
- Added a `read` command class with required path input, optional timeout flag, and usage examples.
- Updated read schema, path utilities, and dispatch to parse selectors from :raw/:L suffixes on path, removing standalone sel usage.
- Changed truncation/error notices to continue with :<nextOffset> and :1 guidance for read and sqlite pagination.
- Added summarizeCode support with tree-sitter summaries, read.summarize settings, and N-API Summary types/exports.
- Added tests and docs updates for path-embedded selectors, summary behavior, and explicit raw/offset SQL/read cases.
- Updated OpenAI completion request construction to include OpenRouter attribution headers and always-on cache directives (1h TTL).
- Updated image-generation requests to replace legacy X-Title with OpenRouter-specific HTTP-Referer, title, and category headers.
- Removed old OpenRouter X-Title defaults from provider options and removed the matching image-gen header assertion test.
- Added session-draft persistence methods in SessionManager to write unsent editor text to an artifacts-sidecar draft file and delete it after single-shot consumption.
- Persisted editor text during interactive shutdown and restored that draft on resume when the editor was empty, enabling Ctrl+D draft recovery.
- Updated Ctrl+D handling in the editor/controller path and added tests covering draft round-trip, artifact cleanup, stale-draft eviction, and in-memory no-op behavior.
- Wrapped hashline execution by merging same-path input sections into one combined section in execution order.
- Applied a new hashline regression test that validates sequential sections against the same file use the original snapshot even when anchors shift beyond rebase limits.
- Clarified anchor operation semantics, including the `+ ANCHOR` insert-after and `< ANCHOR` insert-before forms.
- Added guidance to prefer narrow ops for additions and deletions, reserving `= A..B` for true replacements.
- Reworked examples with an anti-pattern and preferred one-line insert variant to discourage over-broad replacements.
- Added `recordLocalSubmission` and `withLocalSubmission` to replace inline signature set mutations across input and compaction flush paths.
- Signatures are now cleaned up automatically on delivery failure, preventing stale entries from suppressing editor-draft protection on retries.
- Extended test fixtures and added cases covering signature lifecycle for idle, streaming, and fire-and-forget submission paths.
- Added `HindsightSessionState` to `AgentSession` and bound hindsight lifecycle hooks to session state.
- Removed global hindsight state/queue handling and replaced it with per-session `HindsightRetainQueue` batching and scoped flushing.
- Reworked recall, reflect, and retain tools to use `session.getHindsightSessionState()` instead of sessionId-based lookup.
- Updated SDK/task/backend/controller flows to pass `session`/`parentHindsightSessionState`, scope `/memory` behavior, and document it in changelog.
- Added mental-model settings and config defaults for enablement, auto-seed, refresh interval, and render budget.
- Added built-in mental-model seeds and scope-aware rendering with `<mental_models>` extraction, truncation, and tag handling.
- Added `/memory mm` aliases and handlers for list, show, refresh, history, seed, reload, and delete commands.
- Added client APIs and bootstrap/cache wiring so snippets refresh and inject into prompts on startup.
- Added tests covering seed scope behavior, rendering caps, diffs, and backend/session reload behavior.
- Added a `= A..B` hashline rule and adjacent-edge guidance to the tool prompt.
- Added `HashlineReplacementGroup` and `HashlineDeleteEdit` with helpers to find anchor-target lines.
- Added `findReplacementGroup()` and duplicate-boundary absorption to normalize edits with synthetic deletes and warnings.
- Added normalization flow into `applyHashlineEdits` before bucketing, and added duplicate-boundary edge-case tests.
- Updated `CHANGELOG.md` with Added and Fixed unreleased notes for hashline boundary-line behavior.
- Loaded recall, reflect, and retain tool descriptions from new markdown prompt files.
- Updated the retain tool to accept batched `items`, enqueue each memory entry, and return the queued-memory count.
- Reworked retain tests to pass the array payload, assert pluralized queue messages, and confirm single-batch flush behavior.
- Switched HL_EDIT_SEP initialization to read from process.env, trimming the value and defaulting to "~" when missing or invalid.
- Expanded the separator documentation with benchmark data and reliability notes explaining why "~" was selected as the fallback over other candidates.
- Added per-session retain queues with size/time auto-flush, recursive drain, and lifecycle flushes on end/clear/enqueue.
- Changed `hindsight-retain` to validate session state, enqueue writes, and return `Memory queued.` immediately.
- Added `notice` event support and handlers that route error/warning/status messages with source-aware formatting.
- Updated client and tests with shared request mapping, `RequestOptions`, `buildMemoryItem`, and expanded batch/list/doc APIs.
- Added an optional `aliasOf` link and a `pickPrimaryState` helper to track reusable parent Hindsight state.
- Updated task-depth subagent startup to reuse the latest primary state's bank, client, config, tags, and missions while returning silently when no parent exists.
- Skipped alias states in recall-snippet selection and enqueue so auto-recall/retain work only on the parent session.
- Updated tool selection logic to add memory-specific tools when the session backend is set to "hindsight".
- Ensured the new tools "recall", "retain", and "reflect" are appended only if not already requested.
