@lawaty/lacode 1.3.0-beta.9 → 2.0.0-beta.0

package/CHANGELOG.md

@@ -9,6 +9,33 @@ commit as the changelog entry it describes.
 
 ## [Unreleased]
 
+## [2.0.0-beta.0] — The timeline remembers, and the map has one job
+
+### Added
+
+- Durable timeline state: each session's cursor, pending flag, failure count, and next retry time are written beside the timeline JSONL. Pending work resumes after a restart, a successful summary with no bullet text still advances the durable cursor, and deleting a session removes both files.
+- Exponential backoff for failed timeline summarizations. A failure marks the slice pending and schedules a retry; repeated failures grow the wait up to a cap; an idle event during the wait re-arms the timer instead of bypassing it.
+- `timeline.retryMs` and `timeline.maxBackoffMs` options.
+- User stories US-35 (a timeline survives a restart and resumes unfinished work) and US-36 (a failed timeline summary retries with backoff), with story tests in `tests/stories/j-timeline-reliability.test.mjs`.
+
+### Changed
+
+- The `.opencode/context/` map is now five navigation files only. The cartographer no longer creates or maintains `active.md`, the context plugin no longer normalizes a backlog, and `/context-init` and `/context-review` no longer mention one.
+
+### Removed
+
+- `src/lib/active-backlog.ts`, the `.opencode/context/active.md` map file, and its finished-task retention window. The timeline is the single record of work in flight.
+
+## [1.3.1] — reprocess-timeline rebuilds the caller's timeline, and the pool refreshes daily
+
+### Added
+
+- Pool: `scripts/refresh-pool.mjs` rebuilds `~/.config/lacode/pool.txt` from the score output — opencode-go free models first, preserved unscored providers in the middle, the best opencode model last — with an atomic, fail-closed write. `--install-timer` installs a daily systemd user timer; `--from-scores` reuses an existing `model-scores.json`.
+
+### Fixed
+
+- Timeline: `/reprocess-timeline` now runs inline (`subtask: false`) instead of in a child session, so `timeline_reprocess` rebuilds the caller's timeline rather than the child's.
+
 ## [1.3.0-beta.9] — a timeline reply without bullets still lands
 
 ### Fixed
@@ -1105,7 +1132,9 @@ experimental; `latest` remains 0.4.2.
 
 - First published release: pooled subagents over a shared set of models.
 
-[Unreleased]: https://github.com/lawaty/lacode/compare/v1.3.0-beta.9...HEAD
+[Unreleased]: https://github.com/lawaty/lacode/compare/v2.0.0-beta.0...HEAD
+[2.0.0-beta.0]: https://github.com/lawaty/lacode/compare/v1.3.1...v2.0.0-beta.0
+[1.3.1]: https://github.com/lawaty/lacode/releases/tag/v1.3.1
 [1.3.0-beta.9]: https://github.com/lawaty/lacode/releases/tag/v1.3.0-beta.9
 [1.3.0-beta.8]: https://github.com/lawaty/lacode/releases/tag/v1.3.0-beta.8
 [1.3.0-beta.0]: https://github.com/lawaty/lacode/releases/tag/v1.3.0-beta.0

package/README.md

@@ -31,7 +31,6 @@ is also on disk at `node_modules/@lawaty/lacode/CHANGELOG.md`.
 | Distinct notification sounds | A popup says something happened, but not at a glance whether it is a finish, a question or a block. |
 | SSH notification tunnel | Work on a remote host raises its notification on that host, which has no desktop. |
 | Repository context map | Every task starts by exploring your codebase cold, re-deriving the same layout, entry points and boundaries. |
-| Active backlog | What is half-done, and what can be dropped, lives in chat history the next session cannot see. |
 | Standing instructions | The same direction gets repeated every session, or buried in config that is not per-clone. |
 | Image attachments anywhere | A TUI with no image support locks vision models out entirely. |
 
@@ -271,11 +270,9 @@ Agents are stateless between tasks, and exploring your repository is the most
 expensive part of any task — so every agent pays for it again, re-deriving the
 same layout, entry points and boundaries from scratch. LaCode keeps a short,
 maintained map of the repository in `.opencode/context/` — five navigation files
-(`architecture`, `contexts`, `conventions`, `workflows`, `decisions`) plus
-`active.md`, the active backlog, where a finished task stays a week and an open
-one stays until done — and injects a standing rule that every agent reads the
-relevant file before exploring and starts from the canonical entry points it
-lists. The map holds conclusions, not
+(`architecture`, `contexts`, `conventions`, `workflows`, `decisions`) — and injects
+a standing rule that every agent reads the relevant file before exploring and
+starts from the canonical entry points it lists. The map holds conclusions, not
 inventories: it answers "what should I look at first?" and nothing else, which
 is what makes reading it cheaper than the exploration it replaces.
 

package/dist/agents.js

@@ -124,8 +124,6 @@ You own exactly these files and nothing else:
   150 lines.**
 - \`.opencode/context/decisions.md\` \u2014 architectural constraints that bind future
   work. **At most 120 lines.**
-- \`.opencode/context/active.md\` \u2014 work in flight, grouped by the ISO week it was
-  opened. **At most 60 lines.**
 
 The unit is non-blank lines. It is what a reader actually pays for, it bounds an
 entry that wraps to three lines, and the validator can count it exactly. An
@@ -141,8 +139,6 @@ failed run, not a small one.
 
 The plugin applies this on every run. Write entries in the shape that lets it.
 
-- \`active.md\`: a finished task (\`[x]\` plus \`finished YYYY-MM-DD\`) is dropped
-  after 7 days. An open task never expires, however old it is.
 - \`architecture.md\`, \`contexts.md\`, \`conventions.md\`, \`workflows.md\`,
   \`decisions.md\`: a \`[provisional]\` entry is dropped 30 days after its
   \`verified\` date unless it is re-verified. \`[binding]\` and \`[disputed]\` are
@@ -236,8 +232,7 @@ You are mapping a city, not auditing every house. Do not attempt to understand
 every file.
 
 Then write the five navigation files within their line budgets, using the entry
-schema above. Also create \`.opencode/context/active.md\` with its heading and no
-tasks. For each significant subsystem, give a short "Start here" list \u2014 at most
+schema above. For each significant subsystem, give a short "Start here" list \u2014 at most
 about five canonical files, each with a one-phrase reason. This is the single
 most valuable thing you produce: where an agent should open its editor next.
 
@@ -271,37 +266,6 @@ update needed".
 Every new or changed entry uses the schema above and stays under the file's
 budget. If a file is already over budget, consolidate before adding anything.
 
-Also keep \`.opencode/context/active.md\` current: record work that is in flight,
-and mark a task finished \u2014 with the date it finished \u2014 when it is.
-
-# Task backlog
-
-\`.opencode/context/active.md\` holds the work in flight. It is a rolling window,
-not a history: it answers "what is open, and what just finished".
-
-Format \u2014 one line per task, under the ISO week the task was opened in:
-
-    ## 2026-W43
-
-    - [ ] Ship the release notes \u2014 opened 2026-W43
-    - [x] Fix the flaky test \u2014 opened 2026-W43, finished 2026-10-19
-
-The rules:
-
-- A task that is not marked \`[x]\` is OPEN. An open task is never removed,
-  however old it is.
-- A task is FINISHED when it is marked \`[x]\` AND carries a \`finished\` date.
-  Record the date it actually finished, as \`YYYY-MM-DD\`.
-- The week is \`YYYY-Www\`, Monday to Sunday. \`date +%G-W%V\` prints it.
-- A finished task is dropped once it is more than a week old. The plugin
-  enforces that: record the finished task with its date and let the window close
-  on its own. Do not delete old finished tasks by hand, and do not keep a
-  finished task back to preserve it \u2014 the file is meant to stay small.
-- Never invent a task. Record work that actually happened, and mark a task
-  finished only once it is finished.
-- Keep the file inside the window. If the backlog has grown past a handful of
-  weeks, that is a signal to close finished tasks, not to add sections.
-
 # Minimal exploration
 
 Default shape:
@@ -389,7 +353,7 @@ Never drop a binding constraint to make room.
 
 Trigger: an explicit review (see the \`context-review\` command).
 
-Audit the map files and the backlog for:
+Audit the map files for:
 
 - schema \u2014 every entry has one marker, at most three lines, a pointer, and a
   \`verified\` date
@@ -404,8 +368,6 @@ Audit the map files and the backlog for:
 - missing canonical entry points for significant subsystems
 - unsupported assumptions \u2014 anything that is not \`[binding]\` or \`[disputed]\`
   should be \`[provisional]\`
-- backlog tasks that are finished but carry no date \u2014 the retention window
-  cannot age those out, so they stay forever
 
 Fix problems only in the files you own. Never modify source code or a project
 page during a review. Report what you found, what you fixed, what you left as
@@ -877,7 +839,7 @@ if the blocker is understanding rather than execution, report that instead.`,
     }
   },
   "context": {
-    "description": "Repository cartographer and memory maintainer. Bootstraps and incrementally maintains .opencode/context/ (architecture, contexts, conventions, workflows, decisions) as a small navigation map for future agents, and keeps .opencode/context/active.md \u2014 the active backlog of work in flight, grouped by week. Runs automatically after sessions that changed files; can also be invoked with /context-init, /context-update, /context-review. Never use it for general coding tasks.",
+    "description": "Repository cartographer and memory maintainer. Bootstraps and incrementally maintains .opencode/context/ (architecture, contexts, conventions, workflows, decisions) as a small navigation map for future agents. Runs automatically after sessions that changed files; can also be invoked with /context-init, /context-update, /context-review. Never use it for general coding tasks.",
     "mode": "subagent",
     "temperature": 0.1,
     prompt: CARTOGRAPHER_PROMPT,

package/dist/commands.js

@@ -70,8 +70,7 @@ exploration (README \u2192 existing docs \u2192 manifests \u2192 top-level struc
 and framework config \u2192 entry points \u2192 selective source), then write the five
 context files inside their line budgets, using the entry schema (provenance
 marker, rule of at most three lines, \`\u2014 See <path>[ \xA7Section] (verified
-YYYY-MM-DD)\`). Also create \`.opencode/context/active.md\` with its heading and no
-tasks.
+YYYY-MM-DD)\`).
 
 Remember: write conclusions, not inventories. Point at the project's existing
 documentation instead of restating it. Run the drift pass and report a
@@ -117,7 +116,7 @@ Follow the review procedure and entry schema in your system prompt. Everything
 outside \`.opencode/context/\` is read-only for this run. Do not modify source
 code under any circumstances.
 
-Audit the context files and the backlog for:
+Audit the context files for:
 
 - schema \u2014 every entry has one provenance marker, at most three lines, a
   pointer, and a \`(verified YYYY-MM-DD)\` date
@@ -136,8 +135,6 @@ Audit the context files and the backlog for:
 - missing canonical entry points for significant subsystems
 - unsupported assumptions \u2014 anything that is not \`[binding]\` or \`[disputed]\`
   should be \`[provisional]\`
-- backlog tasks that are finished but carry no date \u2014 the retention window
-  cannot age those out, so they stay forever
 
 Fix what you find in \`.opencode/context/\` only, then report: findings, fixes,
 and anything you could not verify.`
@@ -225,9 +222,13 @@ $ARGUMENTS
 
 Report the tool's output as it came back, then add one short line naming what changed: re-admitted, still limited, or benched because the provider does not serve the model. If a model was benched as \`model-missing\`, say so and name the replacement in \`~/.config/lacode/pool.txt\` \u2014 but change nothing yourself.`
   },
+  // `agent: "implement"` is a subagent, so without `subtask: false` this command
+  // would run in a child session and `timeline_reprocess` would rebuild the
+  // child's timeline instead of the caller's.
   "reprocess-timeline": {
     description: "Rebuild this session's timeline from its full transcript",
     agent: "implement",
+    subtask: false,
     template: "Call the `timeline_reprocess` tool exactly once with no arguments. Then reply with only its output."
   },
   "image-in-cb": {