danthamanvoiss Purpose
Document a simple, reusable handoff structure that lets one skill pass only the most relevant context, decisions, and unknowns to the next skill in a chain.
Use when
- One skill's draft or plan will feed another skill.
- You want to reduce repeated prompting across multi-step workflows.
- Outputs need to stay portable between SalesPortl and other harnesses using the same skill suite.
Required inputs
- Source skill name and its completed output or working draft.
- The downstream objective the next skill needs to accomplish.
- Any key decisions, constraints, approvals, or unknowns already identified.
Safety/authority
- Keep the handoff factual and traceable to the source skill output.
- Separate confirmed decisions from open questions or placeholders.
- Do not upgrade authority during handoff; draft-only remains draft-only unless the user explicitly changes it.
Workflow
- Capture the source skill name and the downstream objective.
- Distill the source output into only the decisions, constraints, and assets the next skill actually needs.
- List open questions, approval gaps, and placeholders so the downstream skill does not treat them as settled facts.
- Emit a compact handoff block using the same headings each time.
- When relevant, include which context source was used so downstream skills know whether the input came from harness-native context, root defaults, or a client folder.
Output format
[Skill handoff]
Source skill: ...
Next objective: ...
Context source: ...
Key decisions: ...
Constraints / approvals: ...
Assets or references: ...
Open questions / unknowns: ...
Quality checks
- The handoff is short enough for the next skill to consume quickly.
- Decisions, constraints, and unknowns are clearly separated.
- Source attribution is present so downstream skills can trace the origin.
- The structure is simple markdown guidance, not a schema or executable format.
Related skills
agency-router, context-loader, output-organization-guide, cross-harness-collaboration-notes