Add /stakeholder-summary skill#10
Merged
Merged
Conversation
Turns a feature specification into a plain-language stakeholder summary with Mermaid diagrams for user experience and data flow. Lands the summary next to the source spec so it can be shared with non-technical stakeholders for feedback before implementation kicks off. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replace the single Step 5 self-check with a two-pass structure that partitions attention between lexical scrubbing (Pass A) and structural reading-order checks (Pass B). Each pass requires a fresh Read of the output file from disk so the check runs against actual file contents rather than working memory. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
37fa232 to
51455db
Compare
Four refinements to the stakeholder-summary skill driven by reviewing real output against the spec: - Require a 3-5 sentence prose block under every "after this change" diagram (and under multi-path "Today" diagrams) so a reader who skims the diagrams still learns what each path does and how it differs from the others. The single "Today" diagram keeps its lighter lead-in. - Stop duplicating exclusions into the closing questions. "Not in this slice" now ends with a single catch-all confirmation prompt, and "What we are asking stakeholders" is restricted to genuine trade-offs, framing calls, or unresolved questions from the spec — not yes/no restatements of items already excluded. - Tie diagram counts to the spec, not the template. Both "today" and "after this change" subsections add or drop diagrams to match the number of meaningfully distinct paths the spec describes; the template now models the pattern with annotations instead of locking in 1+2. - Add a new Pass A internal-consistency / contradiction check to Step 5. It catches diagram-vs-exclusion, UX-vs-data-flow, and vocabulary-collision contradictions, resolves them against the source spec when possible, and uses AskUserQuestion with named options and a grounded recommendation when only the user can decide. Existing passes shift to Pass B (plain language) and Pass C (reading order), with cross-references updated.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
/stakeholder-summaryskill: turns a feature specification into a plain-language stakeholder summary with Mermaid diagrams for user experience and before/after data flow. The summary lands next to the source spec.references/stakeholder-summary-template.mdscaffolddocs/skills/stakeholder-summary.mdand updates the skill index,CLAUDE.mdcatalog, and skill-count references.Known divergence from documented standards
docs/guidance/skill-building-guidance/skill-description-frontmatter.mdrequires bidirectional boundary statements: when skill A names skill B in its boundary, B must also name A. This skill's description points at/plan-a-feature,/plan-a-phased-build, and/plan-implementation, but those three descriptions are not being updated to point back here in this PR.Reason: those existing descriptions are already long, and adding another sibling pointer to each one increases token cost paid in every conversation. Skills in this plugin are typically invoked by slash command, so it is plausible the descriptions are doing less trigger-routing work than the rule assumes — and may be context bloat to begin with. A follow-up should either (a) trim the existing descriptions and then add the back-pointers, or (b) revise the bidirectional rule to acknowledge that slash-invoked skills weigh the trade-off differently. Holding off on making that call here.
Test plan
/stakeholder-summaryagainst an existingfeature-specification.mdin a planning folder and confirm the output lands in the same directory.🤖 Generated with Claude Code