Fusion Skills

Failure bug draft: assets/issue-templates/skill-workflow-failure-bug.md.hbs. Before filling, read its .hbs slot header, assets/partials/agent-content/README.md, and every named partial. Fill bug-report templates by hand.

Skill lifecycle entrypoint. Resolve intent, engage the right agent.

Loading behavior

Load ONLY the routed agent file. References are loaded on-demand by the agent when it needs them (e.g., skill-catalog.md only on MCP failure). Do not preload all agents or all references.

Unattended runs: references/service-mode-contract.md defines how a skill runs under a service task (given inputs, allowed actions, stop rules, proposals, fingerprints). Load it only when a skill receives a service task or an author makes a skill service-ready.

MCP tools

This skill uses mcp_fusion_skills — the Fusion MCP tool that handles both skill discovery and lifecycle operations:

• Discovery (intent: query): semantic search over the skills index by description or task.
• Inventory (intent: inventory): list installed skills.
• Lifecycle (intent: install | update | remove): advisory commands for managing skills.

Routing

Skill bug reports from the warden are written for human readers in the fusion-agent-content shape (banner, TL;DR, folded details and evidence). Feedback mode: run feedback. Reports improvements; author mode implements them.

For APM sync or app workflow setup/updates, hand off to fusion-repository-greenkeeping. This skill owns skill lifecycle, not workflow templates or rendering.

If the user was routed here from the fusion main gate with a partially-resolved intent (e.g., "skills"), proceed directly to the most likely agent without re-asking the top-level question.

If intent is genuinely unclear and the user was not already asked a clarifying question, ask one question: "Are you looking to find a skill, install or update one, create your own, or report a problem with one?"

Compound intents: If the request spans multiple agents (e.g., "inspect this skill and fix the issues"), run agents sequentially. Pass the output of the first agent as context to the second.

Loop prevention: If you were activated from the fusion main gate, do not redirect back to fusion. Handle the request within your agents or state that it is out of scope.

Out of scope: Resolving GitHub issues, reviewing PRs, planning task breakdowns, or authoring unrelated GitHub issues. Skill failure/feedback reports belong to warden; direct other work to relevant installed skill or fusion main gate — never re-route back if activated there.

Migration from deprecated skills

If fusion-discover-skills or fusion-skill-self-report-bug is installed alongside this skill, tell the user:

"You have a deprecated skill installed that's now part of fusion-skills. Remove its dependency from apm.yml, then run apm prune to reconcile the lockfile and deployed files."

Repository-policy handoff

Target repo instructions own commits, validation, changesets, PRs, branch refresh and merge policy. Read at point of use; local workflow setup approval never authorizes remote mutations or deployment.

Safety & constraints

This skill is mutation-capable. Repository-local workflow instructions take precedence over inline guidance when they conflict.

• No secrets or credentials.
• No GitHub mutations without confirmation, except run feedback under trusted runtime policy.
• No remote script execution.
• No invented skill names or catalog results.
• Installed skill files are copies (use apm.yml and apm.lock.yaml to inspect dependency provenance and deployed-file ownership) — never edit them in place directly; route content fixes through agents/author.agent.md (which redirects to fusion-skill-authoring, source repo) and failures through agents/warden.agent.md (report mode).