adeelahmad · Oct 18, 2026 · Oct 18, 2026 · Oct 18, 2026
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -50,6 +50,11 @@ tag in lockstep. Tag releases as `vMAJOR.MINOR.PATCH`.
   overrides it (`breach:` line) and acts per class (SKILL.md "Breach classes").
 
 ### Added
+- **`task-worktree merge <TASK_ID> <BRANCH>`** records the §4.7 lineage on Claude Code, where the
+  host makes the worktree and the supervisor merges it (so `task-worktree remove` never runs): a
+  `--no-ff` merge, then `worktree.merged` + one `commit.created` per merged commit. SKILL.md tells the
+  supervisor to merge with it (a raw `git merge` left live runs 11–14 with no traceable commits), and
+  `make e2e-live` now checks every merged commit answers `events trace`.
 - Stuck release: a worker that hits the same gate block 3 times in a row (one gate, task,
   attempt and worktree; `AGENTIC_STUCK_BLOCKS`) is released **unverified** instead of
   looping until its time box (live run 10 lost 29 minutes to a RED worker that stopped 29
@@ -245,6 +250,9 @@ tag in lockstep. Tag releases as `vMAJOR.MINOR.PATCH`.
   reading it stays allowed. 2 checks failed before.
 
 ### Fixed
+- The planner persona states `targets check`'s rules (`{"targets": [...]}`, `symbol` only for code
+  that exists, `planned_symbol` with `create`/`extend` for new code), so its first `targets.json`
+  passes (live runs 12–14: 2–3 blocks each).
 - The planner persona lists the exact frontmatter and `##` sections `planning-artifacts.kdl`
   requires, so its first write passes (every live run so far wrote each artifact twice: 8–12
   write blocks per run).

diff --git a/docs/ROADMAP.md b/docs/ROADMAP.md
@@ -14,7 +14,7 @@ spec; this table is the plan. Versions are targets, not promises.
 | Milestone | Target | Contents | Status |
 |---|---|---|---|
 | **M0 — Hygiene** | v0.10.x | Mark `DESIGN` / `ARCHITECTURE` / `KICKSTART` historical; fix stale counts (8 → 9 roles) and the README CI section; add the `CONTRIBUTING.md` the README badge links to; one live-session smoke test of the hook wiring (open since `DEVLOG` v0.1) | done: live smoke test `make smoke-live YES=1` (6 checks, ~$0.004 on haiku); its first run found the Linux fail-open in `gate-supervisor-scope` (fixed, regression test added) |
-| **M0.8 — North-star run** | v0.11.x | `make e2e-live YES=1`: a one-line intent (semver parser) goes through retrospective, intake, standards, planner, RED/SCAFFOLD/GREEN per task, structural review and the final gate to merged, tested code in a real Claude Code session, with the approval given up front. Reason: nothing had ever run the whole loop live; the first run found that no role gate fired on plugin installs | run 1: done end to end, 19 min, $1.15 (sonnet supervisor, haiku workers), 22 tests on main, 9/11 checks; found and fixed the `aloop:<role>` matcher bug, a selfcheck crash and copy-merging. Run 2: **11/11**, 12m46s, $0.92, host gates on every role, worktree branches merged with `--no-ff`; found gates looping on the supervisor's contract slips (fixed). Run 3: 11/11, 12m00s, **$0.79**; found GREEN looping on RED's unformatted tests (fixed: RED/SCAFFOLD run the format check) and selfcheck missing the contract in Claude Code worktrees (fixed: the contract is copied in). Run 4 (all fixes): 11/11, 17m56s, $1.10, 12 gate blocks and no loops (28 → 17 → 12). Cost varies $0.79–1.15 per run; next: cut it (scaffold symbol parsing slips such as `(enum,`, supervisor turns), then a second intent beyond the semver fixture. Run 5: 11/11 but 29 min, $1.83, 44 blocks (the planner made 5 tasks; 27 false isolation blocks after workers `cd`'d to the main tree, fixed; subagents also read the plugin's own source, which costs tokens). Run 6: 11/11, **11m10s, $0.65**, plugin-source reads 30 → 2 after the agent prompts said not to; found a dispatch-before-contract loop (25 blocks, fixed by `gate-dispatch`) and annotated scaffold-symbol lines (fixed). Run 7 (after #9): 11/11, 13m46s, $0.98, 20 blocks, no dispatch loop; found RED blocked for listing an earlier task's passing tests, scaffold-symbols tables with a header row and trait-impl lines, and a guessed memory.md frontmatter (all fixed; `aloop init` now seeds memory.md). The re-plan coverage loss is now caught by `gate-final` (every planned test must be on main). Run 8, **second intent** (`INTENT=wordfreq`, two dependent stories): 12/12, 18m05s, $1.53, 54 blocks; found scope globs expanded against the worktree (29 GREEN blocks) and `mod common;` across test crates reported as a HIGH duplicate (57 review halts), both fixed. Run 9 (wordfreq again): 12/12, 19m42s, $1.56, 29 blocks; the ctx-symbols fix had not reached the run (`ensure-tools` rebuilt only on a version bump, fixed) and the planner named story dirs `S1-01/` (block message now names `s1-01-<slug>/`). Run 10 (rebuilt tools): 12/12, $1.20, but 44m48s and 45 blocks: the duplicate false positive is gone (0 HIGH); a RED worker's Write/Edit to the story's output.md was refused by Claude Code's worktree isolation and it stopped 29 times without trying a shell append (fixed: the block and the personas say how), plus 3 stale-`ATTEMPT` tamper blocks (fixed). Fixed after run 10: a worker that repeats an identical block 3 times is released unverified with a `STUCK` marker the supervisor acts on (reason: 29 minutes lost in run 10). Run 11 (all run-10 fixes): 12/12, **12m56s, $0.86, 6 blocks** (from 44m48s, $1.20, 45): no repeated block, one RED report-append block fixed on the first retry, one scaffold fmt block, two structural-reviewer comms blocks (STORY_DIR unset for a sprint-wide review, then a stale block) and 12 `gate-validate-artifact` write blocks on the planner's artifacts. Fixed after run 11: the artifact-write block carries md-db's diagnostics (the planner guessed twice per artifact: 12 of 25 verdicts), `gate-dispatch` requires an existing `STORY_DIR` and names the sprint dir for a whole-wave structural review (2 blocks), and OpenCode reports a stuck release as unverified. Run 12 (those fixes): 12/12, 15m22s, $1.10, 21 blocks: the planner fixed each artifact on the first retry (8 write blocks, from 12) and the reviewer's contract had `STORY_DIR`, but a stale worktree checkout under `.claude/worktrees/` made `ctx-symbols conflicts` report every symbol of main as a duplicate and the reviewer was blocked 10 times on a HIGH finding it could not fix (fixed: the walk skips host checkouts, `.agentic/`, `.transcripts/`). The stuck release now covers the structural reviewer and the final gate too (reason: 10 identical blocks, run 12). Gap: `targets check` blocked the planner twice because a context target named a planned symbol not yet in `src/lib.rs` (reason: planned symbols do not exist before SCAFFOLD; §4.6 owner to decide). Run 13 (those fixes): 12/12, 16m31s, $1.33, 16 blocks; the first build of the walk exclusion also excluded a worktree that IS a host checkout, so the scaffold gate counted every symbol 0 times (9 blocks; the worker was then released by the stuck release and the run still finished); fixed before the run ended (the exclusion applies below the walk root) and covered by a conformance case. Open: `targets check` blocks on the planner's first `targets.json` (2 per run, fixed on retry; §4.6 owner), and sub-agents reading the plugin's own source rose to 17 (from 2–10). Run 14 (all fixes): 12/12, **13m47s, $0.88, 11 blocks**, one host-run block in all (RED touched `src/lib.rs`, fixed on retry), no stuck worker; the rest are planning write blocks fixed on the first retry and workers' own self-checks. Next: supervisor-written GREEN instructions overreaching the next task; cut the planner's first-write blocks (its templates should pass the schema as written) |
+| **M0.8 — North-star run** | v0.11.x | `make e2e-live YES=1`: a one-line intent (semver parser) goes through retrospective, intake, standards, planner, RED/SCAFFOLD/GREEN per task, structural review and the final gate to merged, tested code in a real Claude Code session, with the approval given up front. Reason: nothing had ever run the whole loop live; the first run found that no role gate fired on plugin installs | run 1: done end to end, 19 min, $1.15 (sonnet supervisor, haiku workers), 22 tests on main, 9/11 checks; found and fixed the `aloop:<role>` matcher bug, a selfcheck crash and copy-merging. Run 2: **11/11**, 12m46s, $0.92, host gates on every role, worktree branches merged with `--no-ff`; found gates looping on the supervisor's contract slips (fixed). Run 3: 11/11, 12m00s, **$0.79**; found GREEN looping on RED's unformatted tests (fixed: RED/SCAFFOLD run the format check) and selfcheck missing the contract in Claude Code worktrees (fixed: the contract is copied in). Run 4 (all fixes): 11/11, 17m56s, $1.10, 12 gate blocks and no loops (28 → 17 → 12). Cost varies $0.79–1.15 per run; next: cut it (scaffold symbol parsing slips such as `(enum,`, supervisor turns), then a second intent beyond the semver fixture. Run 5: 11/11 but 29 min, $1.83, 44 blocks (the planner made 5 tasks; 27 false isolation blocks after workers `cd`'d to the main tree, fixed; subagents also read the plugin's own source, which costs tokens). Run 6: 11/11, **11m10s, $0.65**, plugin-source reads 30 → 2 after the agent prompts said not to; found a dispatch-before-contract loop (25 blocks, fixed by `gate-dispatch`) and annotated scaffold-symbol lines (fixed). Run 7 (after #9): 11/11, 13m46s, $0.98, 20 blocks, no dispatch loop; found RED blocked for listing an earlier task's passing tests, scaffold-symbols tables with a header row and trait-impl lines, and a guessed memory.md frontmatter (all fixed; `aloop init` now seeds memory.md). The re-plan coverage loss is now caught by `gate-final` (every planned test must be on main). Run 8, **second intent** (`INTENT=wordfreq`, two dependent stories): 12/12, 18m05s, $1.53, 54 blocks; found scope globs expanded against the worktree (29 GREEN blocks) and `mod common;` across test crates reported as a HIGH duplicate (57 review halts), both fixed. Run 9 (wordfreq again): 12/12, 19m42s, $1.56, 29 blocks; the ctx-symbols fix had not reached the run (`ensure-tools` rebuilt only on a version bump, fixed) and the planner named story dirs `S1-01/` (block message now names `s1-01-<slug>/`). Run 10 (rebuilt tools): 12/12, $1.20, but 44m48s and 45 blocks: the duplicate false positive is gone (0 HIGH); a RED worker's Write/Edit to the story's output.md was refused by Claude Code's worktree isolation and it stopped 29 times without trying a shell append (fixed: the block and the personas say how), plus 3 stale-`ATTEMPT` tamper blocks (fixed). Fixed after run 10: a worker that repeats an identical block 3 times is released unverified with a `STUCK` marker the supervisor acts on (reason: 29 minutes lost in run 10). Run 11 (all run-10 fixes): 12/12, **12m56s, $0.86, 6 blocks** (from 44m48s, $1.20, 45): no repeated block, one RED report-append block fixed on the first retry, one scaffold fmt block, two structural-reviewer comms blocks (STORY_DIR unset for a sprint-wide review, then a stale block) and 12 `gate-validate-artifact` write blocks on the planner's artifacts. Fixed after run 11: the artifact-write block carries md-db's diagnostics (the planner guessed twice per artifact: 12 of 25 verdicts), `gate-dispatch` requires an existing `STORY_DIR` and names the sprint dir for a whole-wave structural review (2 blocks), and OpenCode reports a stuck release as unverified. Run 12 (those fixes): 12/12, 15m22s, $1.10, 21 blocks: the planner fixed each artifact on the first retry (8 write blocks, from 12) and the reviewer's contract had `STORY_DIR`, but a stale worktree checkout under `.claude/worktrees/` made `ctx-symbols conflicts` report every symbol of main as a duplicate and the reviewer was blocked 10 times on a HIGH finding it could not fix (fixed: the walk skips host checkouts, `.agentic/`, `.transcripts/`). The stuck release now covers the structural reviewer and the final gate too (reason: 10 identical blocks, run 12). Gap: `targets check` blocked the planner twice because a context target named a planned symbol not yet in `src/lib.rs` (reason: planned symbols do not exist before SCAFFOLD; §4.6 owner to decide). Run 13 (those fixes): 12/12, 16m31s, $1.33, 16 blocks; the first build of the walk exclusion also excluded a worktree that IS a host checkout, so the scaffold gate counted every symbol 0 times (9 blocks; the worker was then released by the stuck release and the run still finished); fixed before the run ended (the exclusion applies below the walk root) and covered by a conformance case. Open: `targets check` blocks on the planner's first `targets.json` (2 per run, fixed on retry; §4.6 owner), and sub-agents reading the plugin's own source rose to 17 (from 2–10). Run 14 (all fixes): 12/12, **13m47s, $0.88, 11 blocks**, one host-run block in all (RED touched `src/lib.rs`, fixed on retry), no stuck worker; the rest are planning write blocks fixed on the first retry and workers' own self-checks. The supervisor's GREEN mandates in runs 11–14 named one symbol and `src/lib.rs only`, and the diff scope (from `tasks.md` or `targets scope`) bounds them anyway, so the "GREEN instructions overreaching" item is closed as not reproduced. The planner persona now carries the schema's artifact shapes and the `targets check` rules (to measure). Next: sub-agents reading the plugin's own source (4–17 reads per run: tokens), and a live check that every commit on main traces to its task and intent (`events trace`, §4.7, never measured live) |
 | **M0.6 — Gates fail closed** | v0.11.x | A gate that crashes (exit 1, e.g. an unbound variable) is a non-blocking hook error on Claude Code, so the action goes through. Make every gate trap unexpected errors into exit 2 with the reason, and add a conformance case per gate. Reason: the smoke run showed one real instance | done: `_failclosed.sh` on all 10 sub-agent gates + supervisor-scope inside a sprint; `scripts/test/fail-closed.test.py` (5 checks). Left fail-open on purpose: `gate-commit-author` and `gate-validate-artifact` outside a sprint, which see every session on the machine |
 | **M0.7 — Supervisor-owned task contract** | v0.11.x | The contract lives in the main tree, `.agentic/state/tasks/<TASK_ID>.env`, and the dispatch prompt's first line names the task; gates and the time box find it from the host's transcript, a worktree `task.env` that disagrees blocks, and sub-agents may not Write/Edit `.agentic/state/` or `task.env`. Settled by a live probe: Claude Code's `isolation: "worktree"` makes `.claude/worktrees/agent-<id>` at dispatch, so a pre-written in-worktree contract was impossible there. Bash writes are guarded too (a command naming the contract with a write verb is blocked; a heuristic, so an obfuscated write still gets through). On hosts with no sub-agent transcript the worktree's `TASK_ID` (or its `.agentic/worktrees/<id>` path) picks the contract | done: `make conformance` (58 today) + `scripts/test/{budget-worktree,supervisor-scope}.test.py` |
 | **M0.5 — Rename to aloop** | v0.11.0 | Plugin, marketplace, skill, CLI (`aloop`, `aloop init`), OpenCode plugin file and docs renamed; `agentic`, `agentic init`, `agentic-init` and `agentic bind` kept as aliases; OpenCode/Codex installs clean up pre-rename files. Data layout and settings keys unchanged (moved to M1). GitHub repo rename is the owner's call | done: `scripts/test/rename-compat.test.py` |

diff --git a/plugin/agents/planner.md b/plugin/agents/planner.md
@@ -115,7 +115,13 @@ Write `targets.json` beside tasks.md when you know what each task changes: one e
 change (`change_ref` C1…, `task`, `operation` create|modify|extend|refactor|replace|delete,
 `path`, `symbol` or `planned_symbol`, a one-line `reason`, `evidence_refs`, `constraints`).
 Planned files stay planned: never create stubs. `targets check <STORY_DIR>` (also run by
-gate-stage2-complete) proves each target against tasks.md and the real tree.
+gate-stage2-complete) proves each target against tasks.md and the real tree. Its rules, so
+the first write passes (live runs 12–14 hit each of these):
+- the file is `{"targets": [...]}`, never a bare list;
+- `symbol` names something that EXISTS in the tree now; a function this sprint will add is a
+  `planned_symbol` — `create` for a new file (the path must not exist yet), `extend` for a new
+  symbol in an existing file (`"operation": "extend", "path": "src/lib.rs", "planned_symbol": "count_words"`);
+- `modify|refactor|replace|delete` need a `symbol` that exists (or `planned_symbol` for `extend`).
 
 ## Hard limits
 - Writes no production code; does not start RED.

diff --git a/plugin/bin/README.md b/plugin/bin/README.md
@@ -50,7 +50,7 @@ ctx-symbols) → WARN + grep fallback; never a silent pass.
 | `install` | `aloop install codex\|opencode [--project DIR\|--global] [--uninstall]` — renders agents/skills/commands (+ the OpenCode plugin from `hosts/opencode/`) from this plugin's sources; `status`; `claude` |
 | `host` | the `HostAdapter` interface (ROADMAP §4.9, `_host.py`): Claude Code is the reference, Codex and OpenCode are adapters. `describe [--host H] [--json]` prints the contract (capabilities, spawn/wait/send/cancel/subscribe vocabulary, compact/rotate, isolation, gate install/enforcement, tool policy, resolved models) · `check` (every adapter complete; gates `native` or `outer`, never warn; in `make test-hosts`) · `worktree create\|remove\|diff TASK`. A capability a host lacks raises `Unsupported`, never a guess. |
 | `host-adapter` | SubagentStart/Stop: register sub-agents (`.agentic/state/agents/`) for the ledger; PreToolUse on Codex: a worker must `aloop bind <worktree>` first, then its commands/patches must stay in it (denied otherwise) |
-| `task-worktree` | `add\|path\|remove <TASK_ID>` — per-task worktree at `.agentic/worktrees/<TASK_ID>` (branch `agentic/<TASK_ID>`) for hosts without built-in isolation |
+| `task-worktree` | `add\|path\|remove <TASK_ID>` — per-task worktree at `.agentic/worktrees/<TASK_ID>` (branch `agentic/<TASK_ID>`) for hosts without built-in isolation; `merge <TASK_ID> <BRANCH>` — `--no-ff` merge of a host-made branch that emits `worktree.merged` + `commit.created` (the lineage on Claude Code) |
 | `bind` | a worker's first command on Codex/OpenCode: validates its worktree (`aloop bind <path>`) |
 | `_usage.py` | token usage from any host's log: Claude transcripts, Codex rollouts (`.jsonl`/`.zst`, cumulative totals → per-call rows), the OpenCode plugin's usage log |
 | `aloop-init` | project setup: `--show` the defaults for the human, `--apply KEY=VALUE…` writes `docs/agents/defaults.md`, the managed `.gitignore` lines and the repo-local git author; `--ensure-gitignore` (SessionStart, worktree-hygiene) re-adds them |