Exécutez n'importe quel Skill dans Manus
en un clic

Exécutez n'importe quel Skill dans Manus en un clic

marshail-promote-assets

Étoiles4

Forks0

Mis à jour9 juillet 2026 à 16:19

Copies MARSHAIL durable assets — both built-in lifecycle assets under `.marshail/{skills,skills-fallback,agents,rules}/` and repo-specific extensions under `.marshail/extensions/{skills,agents,rules}/` — into the project's config-sync source tree (default `.agent-config/`, overridable via `sync.agent_config_dir`), so the next `agent-conf-sync` run fans them out to all tool layouts. Names are preserved as-is: built-ins keep their `marshail-` prefix, extensions keep their `mx-` prefix. Use when the user asks to "promote MARSHAIL skills/agents/rules", "promote MARSHAIL extensions", "copy from .marshail to .agent-config", "wire MARSHAIL assets into the sync source", or similar — and from `marshail-init` for first-time setup.

Installation

Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.

Exécuter dans Manus

Source

crestreach

crestreach/marshail

Ouvrir le dépôt GitHub Voir les dépôts du créateur

Téléchargement

Exécuter dans Manus

Métiers associésSOC

Basé sur la classification professionnelle SOC

Développeurs de logicielsProfessions informatiques et mathématiques·SOC 15-1252

SKILL.md

readonly

Plus depuis ce dépôt

même dépôt

marshail-delegate-to-implement

crestreach/marshail

MARSHAIL Implement stage. Delegate when the user asks to "implement phase X", "work on packet Y", "code this packet", "execute the plan", "start the next implementation cycle", "do the work for the current target", "write the code + tests for this", "run a Dev-QA pass", "close this cycle". Also delegate after the Plan stage (plan approved) to drive each implementation cycle, and after a verifier failure or reviewer fixup loop to resume coding. The subagent picks the cycle target, confirms / deepens the plan, executes (code + tests + Dev-QA where possible), offers the human a chance to inspect and give feedback, and closes the cycle by updating plan status, changelog, and learning files.

2026-07-094

marshail-delegate-to-analysis

crestreach/marshail

MARSHAIL Repo recon / analysis stage. Delegate when the user asks to "analyze the repo for this change", "do recon", "narrow the search surface", "find the likely files / services / contracts", "identify invariants and test seams", "produce repo-recon.md", "what touches X in this codebase?", "where does Y live?", "scope the code areas before planning". Also delegate when change-brief.md exists and the Analysis stage is in scope, or when a planner / implementer flags they don't know enough about an area. The subagent reads the brief and existing knowledge, walks likely paths, and returns a tight repo-recon.md plus a stale-knowledge list — keeping intermediate exploration out of the parent context.

2026-07-094

marshail-delegate-to-architecture

crestreach/marshail

MARSHAIL Architecture / design stage. Delegate when the user asks to "design the solution", "agree the implementation concept", "discuss the architecture", "propose components / modules / APIs / schemas", "produce architecture-notes.md", "draft an ADR for this", "what's the right shape for this?", "compare design tradeoffs", "high-level layout for this change". Also delegate when change-brief.md and repo-recon.md exist and the change is large enough that the solution shape isn't obvious yet. The subagent drives the design conversation, captures rationale, and (when significant) promotes durable decisions to ADRs.

2026-07-094

marshail-delegate-to-driver

crestreach/marshail

Run a whole MARSHAIL change end-to-end. Delegate when the user asks to "run MARSHAIL on this", "do a full MARSHAIL change for X", "orchestrate this change end-to-end", "drive this through MARSHAIL stages", "manage stage transitions for me", "resume this MARSHAIL change", "pick up where the change left off", "I want one orchestrator instead of juggling stages", or any high-level "do the whole thing under MARSHAIL" intent. Also delegate when starting a new change in a MARSHAIL repo (with `.marshail/` initialized) and the user wants per-stage approval gates managed automatically. The subagent agrees scope, detects current stage from on-disk artifacts, dispatches per-stage agents, and surfaces approval gates.

2026-07-094

marshail-delegate-to-help

crestreach/marshail

MARSHAIL on-demand help. Delegate when the user asks "how does MARSHAIL work?", "what does the X stage do?", "which skill / agent should I use next?", "what stage am I in?", "explain the knowledge layer", "how do I write a delivery plan?", "what goes in repo-recon.md?", "how does the config sync fit in?", "what's the difference between marshail-init and marshail-load?", "what's the marshail acronym?", "MARSHAIL?", or any procedural / conceptual question about the MARSHAIL process, the knowledge layer, the available skills / agents, or how MARSHAIL applies to the current change. Also delegate when the caller is unsure which skill / agent to use, or when another skill / agent encounters a MARSHAIL-meta question that would pollute its context. The subagent reads marshail.md and the repo's .marshail/ tree, returns a concise answer, and (when asked) hands off — read-only.

2026-07-094

marshail-delegate-to-intake

crestreach/marshail

MARSHAIL Intake / framing stage. Delegate when the user asks to "frame the change", "write the change brief", "produce change-brief.md", "structure the engineering framing for this change", "list the non-goals and acceptance criteria", "convert the spec into a brief". Also delegate when the Specification stage is finished (specification.md exists) and the Intake stage is in scope, or when the user has a clear prompt but wants a structured brief before analysis/planning. The subagent turns the spec or prompt into change-brief.md (problem, scope, non-goals, acceptance criteria, constraints, rollout expectations).

2026-07-094

name

marshail-promote-assets

description

marshail-promote-assets

MARSHAIL stores its durable assets under .marshail/ in two trees:

Built-in lifecycle assets — shipped with MARSHAIL itself, under .marshail/{skills,skills-fallback,agents,rules}/. Names use the marshail- prefix and keep it when promoted.
Repo-specific extensions — generated by marshail-learner at the Learn stage (or hand-authored), under .marshail/extensions/{skills,agents,rules}/. Names use the mx- prefix at creation time and keep it when promoted.

To make either set available to all AI assistants (Cursor, Claude, Copilot, VS Code, Junie, …), they must be copied into the config-sync source tree (default .agent-config/, overridable via sync.agent_config_dir in config.yml) consumed by .cyncia, and then synced.

This skill performs the copy step only. Run agent-conf-sync (the cyncia sync skill) afterwards to fan the result out to tool-specific layouts.

When to apply

Trigger phrases:

"promote MARSHAIL skills / agents / rules"
"promote MARSHAIL extensions"
"copy MARSHAIL assets into .agent-config"
"wire .marshail/ into the sync source"
"ship the new learner-generated rules / skills"
"sync MARSHAIL assets to tools" (do this skill, then agent-conf-sync)

Also called from marshail-init for the first-time copy after .marshail/ and .agent-config/ are scaffolded.

Do not apply when the user is editing source files in .marshail/ or asking about MARSHAIL stages — that is normal MARSHAIL work.

Inputs

MARSHAIL source dir (default: .marshail/ at the repo root; the current repo's AGENTS.md may pin a different path). Otherwise: ask the user.
Config-sync source dir (default: sync.agent_config_dir from config.yml, which itself defaults to .agent-config/). The user may keep their config-sync source under a different name/path; honor sync.agent_config_dir when set. Otherwise: ask.
Optional explicit overrides from the user prompt (from <path>, into <path>).
Optional scope flag: built-ins-only | extensions-only | all (default).

Source trees walked

This skill walks two source trees inside the MARSHAIL source dir:

Built-ins (marshail- prefix):
- <marshail_src>/skills/<name>/
- <marshail_src>/skills-fallback/<name>/
- <marshail_src>/agents/<name>.md
- <marshail_src>/rules/<name>.md
Extensions (mx- prefix, applied at creation):
- <marshail_src>/extensions/skills/mx-<name>/
- <marshail_src>/extensions/agents/mx-<name>.md
- <marshail_src>/extensions/rules/mx-<name>.md

Extensions does not include a skills-fallback/ subfolder. If a learner-generated skill needs an inline workflow for tools without subagent support, it ships as a single mx-<name>/SKILL.md with the inline content; there is no parallel fallback tree.

Workflow

Resolve paths. Determine MARSHAIL source dir and config-sync source dir as described above (honor sync.agent_config_dir). Use absolute paths.
Sanity-check. Confirm the MARSHAIL source dir exists and contains at least one of skills/, skills-fallback/, agents/, rules/, or extensions/. Confirm the config-sync source dir exists and contains AGENTS.md.
Plan copies. For each source tree (built-ins, extensions) and each asset folder that exists, copy contents into the matching folder under the config-sync source dir, creating the target folder if missing. Names are preserved as-is — no prefixing or renaming happens at copy time. Built-ins stay marshail-*; extensions stay mx-*.

Mappings (built-ins, names kept):
- <marshail_src>/skills/<name>/ → <agent_config>/skills/<name>/ (preserve the inner SKILL.md plus any extra files in the same folder)
- <marshail_src>/skills-fallback/<name>/ → <agent_config>/skills/<name>/ only when the host environment lacks subagent support — see step 3a below. Otherwise skip the fallback tree.
- <marshail_src>/agents/<name>.md → <agent_config>/agents/<name>.md
- <marshail_src>/rules/<name>.md → <agent_config>/rules/<name>.md
Mappings (extensions, names kept):
- <marshail_src>/extensions/skills/mx-<name>/ → <agent_config>/skills/mx-<name>/
- <marshail_src>/extensions/agents/mx-<name>.md → <agent_config>/agents/mx-<name>.md
- <marshail_src>/extensions/rules/mx-<name>.md → <agent_config>/rules/mx-<name>.md

3a. Pick skills/ vs skills-fallback/ per target tool. MARSHAIL ships two parallel built-in skill flavors:

skills/marshail-delegate-to-* — thin wrappers that delegate to the matching subagent in fresh context. Promoted by default (the subagent-capable flavor).
skills-fallback/marshail-* — full inline workflows for layouts that do not support custom subagents. Used as the fallback flavor.

Which flavor is promoted is controlled by sync.skill_flavor in config.yml (default delegate; set to fallback to promote the inline skills instead). Both folders intentionally hold sibling names; promote at most one flavor per name to avoid conflicts. Extensions have no fallback flavor — copy them as-is regardless. 4. Show the plan: list the source files / folders and their destinations (built-ins as marshail-*, extensions as mx-*), then ask for approval before copying. Skip the prompt if the user explicitly said "go ahead", "just do it", or similar. 5. Copy. Use cp -R (Unix) or Copy-Item -Recurse -Force (PowerShell). Overwrite existing files in the config-sync source — that directory is the sync source, not authored content. 6. Skip MARSHAIL housekeeping files. Do not copy the MARSHAIL source dir's own AGENTS.md, ENTRYPOINT.md, LICENSE, config.yml, agent-tools.yml, agent-models.yml, references/, knowledge/, or any README.md (including extensions/README.md). Only the asset folders. 7. Filter by scope or glob (optional). If the user named specific items ("just marshail-plan", "just the new extensions") or passed a scope flag (built-ins-only, extensions-only), narrow the copy plan accordingly. 8. Merge per-agent tool grants. Read agent-tools.yml. For each promoted agent whose canonical name (marshail-<name> for built-ins, mx-<name> for extensions) appears under agents:, union the listed tools into the agent file's declared tools (de-duplicate; never drop the agent's baseline tools). Agents not listed there are copied unchanged. 9. Apply per-agent model assignments. Read agent-models.yml. For each promoted agent whose canonical name appears under agents:, set the listed model as the model: frontmatter key on the promoted copy (add it, or replace an existing value). Agents not listed keep no model: key and run on the host tool's default model. 10. Suggest follow-up. End by suggesting the user run agent-conf-sync (the cyncia sync skill) to propagate the copied assets into tool layouts.

Outputs

Files added or overwritten under <agent_config>/{skills,agents,rules}/.
A short summary listing the copied paths, grouped by source (built-ins / extensions).

Edge cases

Conflicting names. If a non-MARSHAIL skill/agent/rule with the same name exists in the config-sync source, surface the conflict and ask before overwriting.
Out-of-tree MARSHAIL. If the MARSHAIL dir is not at the repo root (rare), the user must specify the path explicitly.
Empty source. If none of skills/, skills-fallback/, agents/, rules/, extensions/ exist in the MARSHAIL source, exit with a note instead of copying.
Misnamed assets. If a built-in is missing its marshail- prefix, or an extension under extensions/ is missing its mx- prefix, surface it and ask — do not silently rename the source.
Dry-run flag. If the user says "preview", "dry run", or similar, print the copy plan without executing.

Pairing

This skill is half of a two-step flow:

marshail-promote-assets — copy .marshail/ assets (built-ins + extensions) into the config-sync source (this skill).
agent-conf-sync — the cyncia sync skill; fans the config-sync source out to all tool layouts.

When the user asks to "promote MARSHAIL assets to all tools", run both in order.