feat(#2505): Phase 4 — runtime-aware subagent dispatch (Option A; resolve-dispatch-type query) (#2525)

* feat(#2508): Phase 4 Option A — runtime-aware subagent dispatch via resolve-dispatch-type query (#2505)

* fix(#2508): prose-variant preamble (avoid scanner-tripping literals) + namedDispatch===false-only mapping

* fix(#2508): remove leftover old-preamble lines (keep prose variant only)

* fix #2508: prose-only reference file

* test #2508: regen golden install parity after workflow preamble additions

* fix #2508: remove preamble from plan-phase.md (Phase 6 capstone ceiling); regen size+golden baselines

* docs(changeset): backfill PR #2525 for Phase 4 (#2508)

docs/INVENTORY-MANIFEST.json

@@ -270,6 +270,7 @@
       "research-verification-protocol.md",
       "reviewer-instances.md",
       "revision-loop.md",
+      "runtime-aware-dispatch.md",
       "scout-codebase.md",
       "security-asvs-levels.md",
       "skeleton-template.md",

docs/INVENTORY.md

@@ -331,6 +331,7 @@ Full roster at `gsd-core/references/*.md`. References are shared knowledge docum
 | `reviewer-instances.md` | Custom reviewer instances for `/gsd-review` (#1517) — same-adapter multi-model review: config shape, resolution rules, invocation, and the REVIEWS.md contract. Lazily loaded by `review.md` when `review.reviewer_instances` is configured. |
 | `universal-anti-patterns.md` | Universal anti-patterns to detect and avoid. |
 | `worktree-branch-check.md` | Canonical spawn-time worktree HEAD/base guard (worktree_branch_check): verify-only and fail-closed — per-agent-branch assertion, protected-ref refusal (#2924), and an exact-base assertion that halts with `exit 42` on mismatch so the orchestrator (worktree lifecycle owner) performs recovery (#48). Embedded into worktree sub-agent prompts at dispatch. |
+| `runtime-aware-dispatch.md` | Runtime-aware subagent dispatch protocol (#2508 Phase 4 Option A): before any `Agent(subagent_type="gsd-*")` call, resolve the type via `gsd_run query resolve-dispatch-type --requested <name> --raw`. On named-dispatch runtimes (Claude/OpenCode/…) the name is returned unchanged; on built-in-only runtimes (kimi-code) it maps to `coder`/`explore`/`plan` by role-suffix. The persona rides `${AGENT_SKILLS_<ROLE>}` (Phase 3) regardless. Documents why a PreToolUse-remap hook (the epic's original Option B) is infeasible — Kimi Code's hook API supports only allow/deny, not tool_input rewriting. |
 | `worktree-path-safety.md` | Worktree guard suite: HEAD assertion, cwd-drift sentinel (step 0a, #3097), and absolute-path guard (step 0b, #3099) — loaded into executor spawn prompts via `<execution_context>`. |
 | `untrusted-input-boundary.md` | Shared prompt-injection boundary (#1577) `@`-included by the 10 research/doc-ingest agents (`gsd-project-researcher`, `gsd-phase-researcher`, `gsd-ui-researcher`, `gsd-assumptions-analyzer`, `gsd-advisor-researcher`, `gsd-doc-classifier`, `gsd-doc-synthesizer`, `gsd-research-synthesizer`, `gsd-ai-researcher`, `gsd-domain-researcher`): treat fetched/read text as data-not-instructions, self-scan before use (PromptArmor 2507.15219), task-anchor (2504.20472), and fence quoted text with a fresh random delimiter per wrap (PPA 2506.05739). Prompt-level defense-in-depth (2503.00061); the hook scanner is a separate pattern pre-filter. |
 | `artifact-types.md` | Planning artifact type definitions. |