git:20260818.c504ab8 to git:20260820.bf0eddd

19 added, 14 removed. Audit A to A.

---

name: longgraph

- description: Author and optionally direct-launch one durable loop-graph run for long-horizon agent work. Use when the user invokes /longgraph, asks to design or start a multi-round run, needs durable state, gated milestones, owner decision boundaries, cross-host continuity, or independent audit. Detect Codex or Claude Code from context instead of asking. For unused/duplicate/slim loops, use /loop-converge. For a self-contained task, recommend the host's ordinary task or goal directly. Do not execute or resume existing runtime node files from this authoring skill.

+ description: Route and author durable long-horizon loop-graph runs. Use for multi-round work needing durable state, gated milestones, owner boundaries, host switching, or independent audit. Route code cleanup to /loop-converge, feature requirements to /loop-deliver, and evidence-backed solution comparison to /loop-research; use loop-graph directly only for a custom run shape. For a self-contained task, recommend the host's ordinary task or goal directly. Do not execute or resume generated runtime node files.

---

- # longgraph — author a loop-graph run

+ # longgraph — route to the right durable run

- longgraph authors through [`loop-graph`](skills/loop-graph/SKILL.md). Specialized

- entries such as [`loop-converge`](skills/loop-converge/SKILL.md) bind a preset

- pack and then follow that skill. Do not execute the generated runtime nodes from

- this authoring skill.

+ longgraph has one shared compiler, [`loop-graph`](skills/loop-graph/SKILL.md),

+ and focused entries that bind only goal-specific rules. All compile the same

+ executor, supervisor, ledger, directives, and ops artifacts. Do not execute the

+ generated runtime nodes from this authoring skill.

- ## Fit check

+ ## Fit and route

- - Use loop-graph when the work needs durable state, many rounds, milestone gates,

- owner boundaries, host switching, or independent verification.

- - If the request is specifically multi-round unused / duplicate / reuse / slim

- work, read and follow [`loop-converge`](skills/loop-converge/SKILL.md) instead

- of designing the North Star from scratch.

+ - Use [`loop-converge`](skills/loop-converge/SKILL.md) for multi-round unused

+ code, duplication, consolidation, reuse, or slimming.

+ - Use [`loop-deliver`](skills/loop-deliver/SKILL.md) for a product or engineering

+ requirement that needs incremental implementation and acceptance proof.

+ - Use [`loop-research`](skills/loop-research/SKILL.md) to compare feasible

+ approaches with open-source implementations, primary research, and controlled

+ experiments before committing to one.

+ - Use [`loop-graph`](skills/loop-graph/SKILL.md) directly only when none of the

+ focused packs fits but the work still needs durable state, gates, owner

+ boundaries, host switching, or independent verification.

- If the request is one self-contained task that fits a normal host goal/task,

explain that longgraph adds no value and recommend sending the task directly. Do

not wrap it in another objective or invent a lighter longgraph mode.

Shared reference: [`lib/methodology.md`](lib/methodology.md) explains why the graph's

- discipline exists. The loop-graph author loads one small per-host reference only after

- the host is selected.

+ discipline exists. The baseline and pack boundary is in

+ [`skills/loop-graph/docs/preset-contract.md`](skills/loop-graph/docs/preset-contract.md).