feat(skills): add project management skill trilogy#52
Merged
Conversation
Adapted from emergent-sports reference implementation, tailored to Finch's architecture layers and label taxonomy. Uses title/label heuristics for component classification until component:* labels are adopted. Closes #49 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Label matches (app-independence, github_actions) now take precedence over title keywords to prevent misclassification of app-layer issues that mention core domain terms. Added missing keywords (account, debt, interest, reflection, recipe, cash flow). Removed ambiguous "tool" keyword from mcp — requires explicit "mcp" match. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adapted from emergent-sports reference, tailored to Finch's architecture and conventions. Reuses the two-pass component classification heuristic from /project-summary. Omits deferred issue handling and tracking issue logic (Finch has no equivalent labels). Uses substantive activity dates for staleness instead of updatedAt. Closes #50 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add developer-experience label → dx component mapping in Pass 1 of both skills, reducing the unclassified bucket. Refine the decomposition heuristic in backlog-review to exclude checklists under acceptance criteria headings, which are verification steps rather than work items needing sub-issues. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adapted from emergent-sports reference, tailored to Finch's architecture and conventions. Uses component dependency graph (core → api → mcp/app) for gate assessment instead of critical-path labels. Phased execution commands with per-phase approval gates ensure no autonomous state changes. Omits deferred issue handling and tracking issue logic (no equivalent labels). Closes #51 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Cold-start mode was skipping both Steps 3 and 4, but Step 5b depends on the gate assessment from Step 4. The gate assessment is independent of whether a milestone exists — it looks at open issues by component in dependency order. Only Step 3 (completion assessment) should be skipped in cold-start mode. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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
1 participant
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.
Overview
The purpose of this change is to add three Claude Code skills that provide structured project management — session orientation, backlog grooming, and milestone planning. This represents an incremental improvement that replaces ad-hoc
gh issue listexploration with repeatable, layered workflows adapted from the emergent-sports project's skill hierarchy./project-summary/backlog-review/milestone-lifecycleAll three skills share a two-pass component classification heuristic (label match first, then title keywords) that maps issues to Finch's architectural layers (core, api, mcp, app, infra, dx). Each was tested by invocation during development, with refinements applied based on observed classification accuracy and structural gaps.
Key design decisions
app-independence,github_actions,developer-experience) take precedence over title keyword matching to prevent misclassification when titles mention cross-cutting termscritical-pathlabelsdeferred,epic,handoff)/milestone-lifecycleproposesghcommands but never auto-executes — every mutation requires explicit user confirmationCloses #49
Closes #50
Closes #51
🤖 Generated with Claude Code