Meta-skill for orchestrating end-to-end feature development from persona research through deployed, tested code — coordinating skills, commands, and agents across every stage. Use when building a new feature end-to-end, coordinating frontend + backend work, or needing a structured approach from research to production.
Security Analysis
medium confidenceThis is an instruction-only meta-skill that routes work to other development skills and does not itself request credentials or install code, but you should verify the referenced skills/commands and the (unpublished) source before trusting autonomous runs.
The skill's name/description (an orchestrator/meta-skill) matches what the SKILL.md actually does: it coordinates other skills, commands, and agents through an eight-stage workflow. It does not request unrelated binaries, env vars, or credentials, which is consistent for a coordinator/meta-skill.
Runtime instructions are limited to routing to other internal commands/skills (e.g., /create-persona, /create-component, /new-feature) and enforcing stage gates. The SKILL.md does not instruct the agent to read unrelated files, exfiltrate data, or call external endpoints directly. It does assume the presence of other skills/commands which will perform concrete actions.
There is no install spec in the registry (instruction-only), so nothing is written to disk by the registry entry itself. README includes manual install examples and an npx command referencing a GitHub path; following those external install instructions would fetch code from outside the registry — review the external repo before running such commands.
The skill declares no required environment variables, credentials, or config paths. That is proportionate given that the skill only orchestrates other internal skills. However, the security posture depends on the privileges of the referenced skills/commands it will invoke.
always:false and normal model invocation are set (defaults). The skill does not request permanent presence or system-wide configuration changes. Note: as a meta-skill it will invoke other skills/agents — those invoked skills may have different persistence/privilege profiles that you should review.
Guidance
This skill is internally consistent for a meta-orchestrator: it only routes to other skills and doesn't ask for secrets or install code itself. Before installing or running it, do the following: (1) confirm the source repository or homepage (registry lists source as unknown); (2) inspect the SKILL.md / SKILL files of the referenced commands and skills (/create-persona, /create-component, api-development, testing-workflow, etc.) to see what those actually do and what credentials they require; (3) avoid running the README's npx/git copy commands unless you trust the external GitHub repo; and (4) if you plan to allow autonomous invocation, ensure the invoked skills have appropriate, minimal permissions and do not contact untrusted external endpoints. If you want higher assurance, provide the source repo or the SKILL.md files for the referenced skills so they can be reviewed as well.
Latest Release
v1.0.0
Initial release of the "full-stack-feature" meta-skill for orchestrating end-to-end feature development. - Enables coordinated, stage-gated workflows spanning persona research, spec writing, UX/UI, API, frontend, backend, testing, and documentation. - Provides an eight-stage feature pipeline, each with defined outputs, commands, and gate checks to ensure nothing ships half-baked. - Includes a routing table for delegating steps to the right skill, command, or agent. - Details a vertical slice development strategy for incremental delivery. - Offers complexity guidance to adapt the process for trivial to epic features.
More by @wpank
Published by @wpank on ClawHub
