loss-driven-development

Documented sugar for LDD[level=L3]: — runs the 5-phase design protocol (Constraints → Non-goals → 3 Candidates → Scoring+Dialectical → Deliverable) from skills/architect-mode/SKILL.md by setting the thinking level to L3 for the next task. Opt-in, not default. Accepts optional creativity argument — /ldd-architect [conservative|standard|inventive]. This command does NOT introduce a separate `mode` axis; it is equivalent to the user typing `LDD[level=L3]:` (or `LDD[level=L3, creativity=<value>]:`) on the next task.

Pre-grant, list, check, or revoke autonomy acks so architect-mode's `creativity=inventive` can fire in autonomous runs (loops, cron, remote agents) without a synchronous human ack. HMAC-signed, scoped to a task family, short-TTL by default.

Show the current LDD hyperparameter config — which layer each value comes from (bundle default / project file / session override / inline) with full provenance.

Explain why LDD just invoked the skill(s) it invoked — cite the trigger phrase from the user's message that matched the dispatch table.

Explicitly install or re-install the LDD plugin's project-local artifacts (.ldd/ldd_trace launcher, statusline, heartbeat + stop-render hooks, settings merge). Idempotent; version-gated; safe to re-run. Required once per project as the opt-in step — unless one of the auto-opt-in signals (plugin self-identification, user-global flag) already fires; afterwards the plugin's SessionStart hook maintains everything automatically on every session start and every plugin update.

Apply after any non-trivial implementation task — evaluates whether an ADR is warranted, writes to Atelier-ADR repo when yes, names the skip reason when no, and generates E2E tests for every new structural invariant so LDD has a loss signal. High bar: most tasks produce no ADR.

The 5-phase design protocol that is active at thinking-levels L3 and L4. NOT a separate mode — it is the discipline the level preset pulls in. Opt-in via four paths — inline `LDD[level=L3]:` (or `LDD[level=L4]:`) prefix, the `/ldd-architect` command, trigger phrases like "design" / "architect" / "from scratch" / "greenfield", or the thinking-levels auto-dispatch landing at L3 / L4 (9-signal scorer; see `../using-ldd/SKILL.md` § Auto-dispatch: thinking-levels). Supports three creativity levels (`conservative` | `standard` | `inventive`) that change the loss function, not the amount of structure.

Use at the start of any LDD task when there is no existing `.ldd/trace.log` readable in the working directory — fires before any other LDD skill writes trace state. Detects the host's persistence capabilities (filesystem, artifact/canvas, conversation-history, memory-pointer, inline-only), picks the most durable tier silently without prompting the user, restores any prior trace state it finds, and announces the chosen store in the LDD trace block's `Store :` header line. Makes LDD portable across Claude Code, Claude Desktop, Codex, Gemini CLI, Cursor, ChatGPT, and any other host the LLM can introspect.

Use when the agent or the user wants to introduce a new measurable quantity as a first-class loss component — e.g., tracking cyclomatic complexity alongside test_pass_rate, or adding a domain-specific rubric to a reasoning chain. Forbids ad-hoc arithmetic on observations and forbids using a new metric as a load-bearing decision gate until calibration has passed. Extends LDD's loss framework to agent-defined objectives without touching LDD core.

Use when a reasoning task requires a multi-step Chain-of-Thought (CoT) AND the task is verifiable (grounded answer, testable code, derivable proof). Forbids greedy step-by-step commitment — every step in the chain goes through thesis/antithesis/synthesis BEFORE entering the chain, using memory-primed antitheses to attack the proposed step. Produces a calibrated predicted-correctness per step and per chain. Applies SGD-on-thoughts in the same way `dialectical-reasoning` applies SGD-on-actions.

1 hook across 1 event