Agent skill
generate-epics
Translate product intent or a PRD into a small set of outcome-driven Epics (3–7 max). Activate when starting a new product, adding a significant feature domain, or breaking down a PRD into actionable user outcomes.
Install this agent skill to your Project
npx add-skill https://github.com/Fr-e-d/GAAI-framework/tree/main/.gaai/core/skills/discovery/generate-epics
Metadata
Additional technical details for this skill
- id
- SKILL-GENERATE-EPICS-001
- track
- discovery
- author
- gaai-framework
- status
- stable
- version
- 1.0
- category
- discovery
- updated at
- 1769472000
SKILL.md
Generate Epics
Purpose / When to Activate
Activate when:
- Starting a new product
- Adding a significant feature or domain
- Restructuring product scope
- Breaking down a PRD into actionable outcomes
Works with or without a PRD.
Process
-
Read the Epic template at
contexts/artefacts/epics/_template.epic.mdbefore writing any Epic file.CRITICAL — ID Collision Guard (MUST execute before assigning any Epic ID):
- a) Scan
contexts/backlog/active.backlog.yamlto find the highest existing Epic number (e.g., if E52 is the last, the next Epic must be E53 or higher). - b) Also scan
contexts/artefacts/epics/for existing.epic.mdfiles to catch any that may not yet be in the backlog. - c) For each Epic file to be created, check if the file already exists at
contexts/artefacts/epics/{id}.epic.md. If it exists with different content, STOP immediately — surface the conflict to the human. - d) The new Epic ID =
max(existing IDs) + 1. Never reuse an Epic ID, even if the previous Epic was deleted or superseded. - Rationale: In a past incident, two concurrent sessions both assigned the same Epic ID to different epics. The second overwrote the first's stories. This guard prevents recurrence.
- a) Scan
-
Think in user outcomes, not features
-
Keep Epics high-level and value-focused
-
Avoid implementation detail
-
Limit to 3–7 Epics maximum
-
For each Epic, answer: "What meaningful user result will this create?"
-
Set domain based on the Epic's primary intent (e.g., engineering, marketing, legal). Leave empty if not applicable.
-
Output using the canonical Epic template
Outputs
Template: contexts/artefacts/epics/_template.epic.md
Produces files at contexts/artefacts/epics/{id}.epic.md.
Key sections per Epic:
- Purpose: what user outcome this delivers and why it matters
- Scope: high-level description of what is included
- Out of Scope: what is explicitly excluded
- Stories: list of story IDs (descriptive only; authoritative tracking is in the backlog)
- Success Metrics: how to know the Epic delivered value
- Dependencies: other Epics or external factors this depends on
Quality Checks
- Each Epic expresses a user outcome, not a technical feature
- Maximum 7 Epics per initiative
- No implementation detail present
- Each Epic is independently valuable
- No Epic ID collision — the assigned ID does not exist in the backlog or on disk with different content
Non-Goals
This skill must NOT:
- Generate Stories (use
generate-stories) - Make technical architecture decisions
- Produce more than 7 Epics per initiative
Epics are the bridge between vision and execution.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
ci-watch-and-fix
Watch GitHub Actions CI after PR creation, detect failures, extract logs, apply minimal fixes, and re-push — keeping the delivery session alive until CI resolves or escalating after 3 cycles. Activate immediately after gh pr create and before marking the story done.
qa-review
Validate that implemented code fully satisfies Story acceptance criteria, respects rules, and introduces no regressions. This is the hard quality gate — no pass means no delivery. Activate after implementation is complete.
compose-team
Assemble the context bundles for each sub-agent based on evaluate-story output. Produces spawn-ready packages for Planning, Implementation, QA, or MicroDelivery sub-agents. Activate after evaluate-story, before spawning any sub-agent.
coordinate-handoffs
Validate sub-agent handoff artefacts, sequence phase transitions, and manage retry and escalation logic. Activate after each sub-agent terminates to determine next action.
implement
Generate correct, minimal, maintainable code that satisfies a validated Story's acceptance criteria against an execution plan. Activate when a Story is validated, a plan exists, and all prerequisites are unambiguous.
delivery-high-level-plan
Transform validated Stories into a clear, minimal, governed execution plan. Used by the Planning Sub-Agent as the first planning pass before prepare-execution-plan for Tier 2/3, or as the sole planning output for simple Stories.
Didn't find tool you were looking for?