1. Isolate `render_diff_report` tests into tmpdir — flaky under `cargo test --workspace`; reads real working-tree state; breaks CI during active worktree ops
2. Expand GitHub CI from single-crate coverage to workspace-grade verification — current `rust-ci.yml` runs `cargo fmt` and `cargo test -p rusty-claude-cli`, but misses broader `cargo test --workspace` coverage that already passes locally
3. Add release-grade binary workflow — repo has a Rust CLI and release intent, but no GitHub Actions path that builds tagged artifacts / checks release packaging before a publish step
4. Add container-first test/run docs — runtime detects Docker/Podman/container state, but docs do not show a canonical container workflow for `cargo test --workspace`, binary execution, or bind-mounted repo usage
5. Surface `doctor` / preflight diagnostics in onboarding docs and help — the CLI already has setup-diagnosis commands and branch preflight machinery, but they are not prominent enough in README/USAGE, so new users still ask manual setup questions instead of running a built-in health check first
6. Automate branding/source-of-truth residue checks in CI — the manual doc pass is done, but badges, Discord invites, copied repo URLs, and org-name drift can regress unless a cheap lint/check keeps README/docs aligned with `ultraworkers/claw-code`
7. Eliminate warning spam from first-run help/build path — `cargo run -p rusty-claude-cli -- --help` currently prints a wall of compile warnings before the actual help text, which pollutes the first-touch UX and hides the product surface behind unrelated noise
8. Promote `doctor` from slash-only to top-level CLI entrypoint — users naturally try `claw doctor`, but today it errors and tells them to enter a REPL or resume path first; healthcheck flows should be callable directly from the shell
9. Make machine-readable status commands actually machine-readable — `status` and `sandbox` accept the global `--output-format json` flag path, but currently still render prose tables, which breaks shell automation and agent-friendly health polling
10. Unify legacy config/skill namespaces in user-facing output — `skills` currently surfaces mixed project roots like `.codex` and `.claude`, which leaks historical layers into the current product and makes it unclear which config namespace is canonical
11. Honor JSON output on inventory commands like `skills` and `mcp` — these are exactly the commands agents and shell scripts want to inspect programmatically, but `--output-format json` still yields prose, forcing text scraping where structured inventory should exist
12. Audit `--output-format` contract across the whole CLI surface — current behavior is inconsistent by subcommand, so agents cannot trust the global flag without command-by-command probing; the format contract itself needs to become deterministic
4. Wire `SummaryCompressor` into the lane event pipeline — **done**: `compress_summary_text()` feeds into `LaneEvent::Finished` detail field in `tools/src/lib.rs`
12. Lane board / machine-readable status API — **done**: Lane completion hardening + `LaneContext::completed` auto-detection + MCP degraded reporting surface machine-readable state
16.**Commit provenance / worktree-aware push events** — clawhip build stream shows duplicate-looking commit messages and worktree-originated pushes without clear supersession indicators; add worktree/branch metadata to push events and de-dup superseded commits in build stream display
17.**Orphaned module integration audit** — `session_control` is `pub mod` exported from `runtime` but has zero consumers across the entire workspace (no import, no call site outside its own file). `trust_resolver` types are re-exported from `lib.rs` but never instantiated outside unit tests. These modules implement core clawability contracts (session management, trust resolution) that are structurally dead — built but not wired into the CLI or tools crate. **Action:** audit all `pub mod` / `pub use` exports from `runtime` for actual call sites; either wire orphaned modules into the real execution path or demote to `pub(crate)` / `cfg(test)` to prevent false clawability surface.
18.**Context-window preflight gap** — claw-code auto-compacts only after cumulative input crosses a static `100_000`-token threshold, while provider requests derive `max_tokens` from a naive model-name heuristic (`opus` => 32k, else 64k) and do not appear to preflight `estimated_prompt_tokens + requested_output_tokens` against the selected model’s actual context window. Result: giant sessions can be sent upstream and fail hard with provider-side `input_exceeds_context_by_*` errors instead of local preflight compaction/rejection. **Action:** add a model-context registry + request-size preflight before provider call; if projected request exceeds context, emit a structured `context_window_blocked` event and auto-compact or force `/compact` before retry.
19.**Subcommand help falls through into runtime/API path** — direct dogfood shows `./target/debug/claw doctor --help` and `./target/debug/claw status --help` do not render local subcommand help. Instead they enter the request path, show `🦀 Thinking...`, then fail with `api returned 500 ... auth_unavailable: no auth available`. Help/usage surfaces must be pure local parsing and never require auth or provider reachability. **Action:** fix argv dispatch so `<subcommand> --help` is intercepted before runtime startup/API client initialization; add regression tests for `doctor --help`, `status --help`, and similar local-info commands.
20.**Session state classification gap (working vs blocked vs finished vs truly stale)** — dogfooding with 14 parallel tmux/OMX lanes exposed that text-idle stale detection is far too coarse. Sessions were repeatedly flagged as stale even when they were already **finished/reportable** (P0.9, P0.10, P2.18, P2.19), **working but quiet** (doc/branding/audit passes), or **blocked on a specific recoverable state** (background terminal still running, cherry-pick conflict, MCP startup noise, transport interruption after partial progress). **Action:** add explicit machine states above prose scraping such as `working`, `blocked_background_job`, `blocked_merge_conflict`, `degraded_mcp`, `interrupted_transport`, `finished_pending_report`, `finished_cleanable`, and `truly_idle`; update clawhip/session monitoring so quiet work is not paged as stale and completed sessions can auto-report + auto-clean.
