mirror of
https://github.com/magnus919/agent-skills.git
synced 2026-09-11 19:47:12 +03:00
Magnus Hedemarkandfactory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
c0c7690724
Move the 8 directories under bundles/ to the repo root via git mv and remove the now-empty bundles/ directory. Replace the "bundles" entry in pyproject.toml [tool.deptry] extend_exclude with the 8 moved dir names so the moved trees stay excluded from Python dependency analysis. Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
73 lines
2.3 KiB
YAML
73 lines
2.3 KiB
YAML
# Bundle manifest (bundle-manifest-v1) — see schemas/bundle-manifest-v1.schema.json
|
|
# and docs/bundle-manifest-design.md. Paths are relative to this bundle dir.
|
|
|
|
schema_version: 1
|
|
bundle_name: workflow-architect
|
|
|
|
purpose: >-
|
|
Discover a user's actual workflow through active interrogation or passive
|
|
observation, then generate a tailored Agent Skills bundle that encodes it as
|
|
loadable skills with trigger conditions, a decision map, a manifest, and an
|
|
optional kanban board — turning a session that feels aimless into structure.
|
|
|
|
audience: >-
|
|
Users and agents that want to formalize how they work — understand their own
|
|
process, encode it as trigger-driven skills they can reuse, or share it with
|
|
collaborators. Not for single factual questions or tasks already owned by a
|
|
more specific skill.
|
|
|
|
stages:
|
|
- name: Active interview
|
|
skills:
|
|
- skills/interviewer/SKILL.md
|
|
- name: Passive observation
|
|
skills:
|
|
- skills/observer/SKILL.md
|
|
- name: Synthesis and generation
|
|
skills:
|
|
- skills/bundle-builder/SKILL.md
|
|
|
|
included_skills:
|
|
- skills/interviewer/SKILL.md
|
|
- skills/observer/SKILL.md
|
|
- skills/bundle-builder/SKILL.md
|
|
|
|
prerequisites:
|
|
- artifact: Session context with enough turns for observation (or user answers for interview)
|
|
skill: skills/observer/SKILL.md
|
|
- artifact: Convergence on workflow phases, branching signals, and tool preferences
|
|
skill: skills/interviewer/SKILL.md
|
|
|
|
outputs:
|
|
- skills-bundle
|
|
- decision-map
|
|
- manifest
|
|
- kanban-board
|
|
|
|
handoffs:
|
|
- to: user
|
|
artifact: skills-bundle
|
|
note: >-
|
|
The generated bundle is written to the host skill directory and
|
|
registered so the umbrella and sub-skills load on trigger phrases.
|
|
- to: user
|
|
artifact: decision-map
|
|
note: >-
|
|
The Mermaid decision map visualizes workflow phases and branching
|
|
signals for review and sharing.
|
|
- to: user
|
|
artifact: manifest
|
|
note: >-
|
|
The generated manifest maps skill names to trigger conditions, entry
|
|
points, and transition signals for the bundle.
|
|
- to: user
|
|
artifact: kanban-board
|
|
note: >-
|
|
The kanban board is included only when the workflow follows a predictable
|
|
linear path where WIP limits and lane transitions add value.
|
|
|
|
conflicts: []
|
|
|
|
eval_suite:
|
|
- evals/evals.json
|