All authors
Claude Skills by aflowai
github.com/aflowai13 skills0 installs33 views
- CheckPre-commit quality checks — auto-fix formatting and lint, then validate build, typecheck, and tests pass. Run before committing.Votes: 0GitHub stars: 2
- Flow BuildingBackground knowledge for building, editing, and debugging Aflow flow definitions and agent configurations. Covers flow schema, step definitions, state variables, agent turn lifecycle, validation pipeline, and input contracts.Votes: 0GitHub stars: 2
- Operation DevBackground knowledge for adding or modifying platform operations — covers operation IDs, registry entries, executor handlers, compact catalog format, and the full development checklist.Votes: 0GitHub stars: 2
- Test McpGuide end-to-end testing of platform changes via the local MCP server (aflow-local). Covers prerequisites, testing patterns per change type, result interpretation, and feedback collection.Votes: 0GitHub stars: 2
- CheckPre-commit quality checks — auto-fix formatting and lint, then validate build, typecheck, and tests pass. Run before committing.Votes: 0GitHub stars: 2
- Cybernetic Skill AuthoringBackground knowledge for designing, authoring, and maintaining cybernetic skills — the bundled (workflow + manifest + eval suite + activation) capability units used by the cybernetic agent. Covers the Helmsman/Runner/Coach/Driver ensemble, system agents and system skills, the platform-artifact registry, the compose-skill / bind-capability authoring path, StagedChange ratification, task context specs, typed step roles, and the skill-shop / catalog import path. For raw workflow schema, validati...Votes: 0GitHub stars: 2
- Drive Skill DogfoodDrive cybernetic skill execution end-to-end via the aflow-local MCP as operator + product engineer — deliver a real feature through the platform's OWN skills (open-pr-from-request, review-pull-request, pr-shepherd) while the gap-log (where skills/tasks/evaluators/guidance/MCP fall short) is the deliverable. Use when asked to "drive a feature via the skills", dogfood the coding lane, or stress-test skill execution.Votes: 0GitHub stars: 2
- Flow BuildingBackground knowledge for building, editing, and debugging the workflow substrate — the graph of steps, transitions, and state variables shared by classic agent definitions and cybernetic skills. Covers schema, validation pipeline, agent-turn lifecycle for classic agents, and input contracts. For cybernetic skill bundles (manifest + evals + activation + system skills + skill-shop), pair with the `cybernetic-skill-authoring` skill.Votes: 0GitHub stars: 2
- Operation DevBackground knowledge for adding or modifying platform operations — covers operation IDs, registry entries, executor handlers, compact catalog format, and the full development checklist.Votes: 0GitHub stars: 2
- Test McpGuide end-to-end testing of platform changes via the local MCP server (aflow-local). Covers prerequisites, testing patterns per change type, result interpretation, and feedback collection.Votes: 0GitHub stars: 2
- Aflow LoopDevelop Aflow through its own loop from a terminal session — plan nodes on the board, Helmsman conversations that commission, review and publish, the merge under the folder's merge posture, then the post-merge cycle on the operator's checkout and live stack. Use when the operator asks you to drive plan work, run a stream, land a fix through the loop, or close out a merge. Builds on aflow-operate. Not for a commissioned change.Votes: 0GitHub stars: 2
- Aflow OperateOperate a running Aflow stack from a terminal session on the operator's behalf — talk to a Helmsman, start and watch skill runs, read a run's tasks and payloads, resolve or decline a gate, read the plan board, debug a stuck run, judge an output. Use when the operator asks you to drive, watch, unblock or evaluate work in Aflow (local edition or hosted), whatever the work is about. Not for a commissioned change.Votes: 0GitHub stars: 2
- Plans**Status:** 🌱 design. Nothing built. **Builds on:** Plan 322 (the plan tree and its operations; the stream that wakes itself, `plan.node.next`; D3 — `plan.node.*` are the Helmsman's, never a skill's and never a Runner's), Plan 328 (a harness is a Runner with a coding agent; §5 P3 — this plan's planning skill uses the harness as its investigator). **Feeds:** the planning skill, the board, and Plan 331 D6, whose wake answers `recalibrate` with this skill.Votes: 0GitHub stars: 2