-
-
Notifications
You must be signed in to change notification settings - Fork 2.4k
Add problem-solving skills from amplifier patterns #1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
**Mode-switching enhancements:** - Update brainstorming skill with flexible UNDERSTANDING/EXPLORING/DESIGNING modes - Update executing-plans skill with REVIEWING/EXECUTING/BLOCKED/REPORTING modes - Add inline dot diagrams with full contextual labels for all valid transitions - Enforce explicit mode switch announcements **New architecture skills from amplifier patterns:** - preserving-productive-tensions: Recognize when to preserve multiple valid approaches vs force resolution - tracing-knowledge-lineages: Understand idea evolution to avoid repeating failures - detecting-emergent-patterns: Cross-domain synthesis, collision-zone thinking, simplification cascades All skills pressure-tested with subagents using RED-GREEN-REFACTOR methodology.
- Reference preserving-productive-tensions from brainstorming EXPLORING mode - Reference detecting-emergent-patterns from brainstorming when stuck - Reference tracing-knowledge-lineages from brainstorming before changes - Reference tracing-knowledge-lineages from executing-plans REVIEWING mode Makes new architecture skills discoverable through commonly-used workflows.
Replace 'ai_working/decisions/' with generic decision record locations
Use common locations instead of project-specific paths
**Restructure:** - Split detecting-emergent-patterns (216 lines) into 5 focused skills (~60 lines each) - Move tracing-lineages to skills/research/ (better category fit) - Create skills/problem-solving/ category **New problem-solving skills:** - simplification-cascades: Find one insight that eliminates multiple components - collision-zone-thinking: Force unrelated concepts together for breakthroughs - meta-pattern-recognition: Spot patterns in 3+ domains - inversion-exercise: Flip assumptions to reveal alternatives - scale-game: Test at extremes to find fundamental limits - when-stuck: Dispatch skill that maps stuck-symptoms to techniques **Improvements:** - Symptom-based when_to_use (what you'd actually think/feel) - Quick reference tables for rapid scanning - Pattern-based triggers (not exact phrase matching) - Cross-references from brainstorming for discoverability Each skill now: scannable in 30 seconds, applicable immediately, focused on one technique.
**Discovery enhancement:** - Add Mandatory Workflow 4 to getting-started: "When Stuck" - Creates path: getting-started → when-stuck dispatch → specific technique - Prevents spinning wheels without seeking help **Trigger clarity:** - Replace phrase examples with pattern descriptions - "Partner reveals constraint you didn't consider" (pattern) not "Oh, I forgot..." (phrase) - Emphasizes NEW information that invalidates current context - Prevents ambiguity between new constraints vs clarifications Tested: All skills verified with subagents, triggers work correctly.
Agents don't search for skills when stuck, even with explicit workflow. Trust organic discovery via symptom-based when_to_use instead.
Mode-switching was overcomplicated. Back to phases with explicit permission to revisit earlier phases when constraints emerge or validation fails. Simpler, clearer, preserves what worked.
Credit amplifier repo (github.com/microsoft/amplifier @ 2adb63f) for agent patterns that inspired these skills.
WalkthroughAdds new documentation skills and attribution ABOUT pages across architecture, problem-solving, and research domains; updates two collaboration skills with version bumps and new guidance on revisiting steps and asking for help. Introduces multiple new problem-solving techniques and a dispatch guide; adds a research lineage-tracing skill and an architecture skill on preserving productive tensions. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
actor Practitioner
participant Dispatch as When-Stuck Dispatch
participant Technique as Selected Technique
participant Output as Outcome
Practitioner->>Dispatch: Describe stuck symptom
Dispatch->>Dispatch: Map symptom to technique
Dispatch-->>Practitioner: Recommend specific skill
Practitioner->>Technique: Apply process (steps/examples)
alt Resolved
Technique-->>Output: Document insights/actions
Output-->>Practitioner: Next steps
else Still stuck
Technique-->>Dispatch: Feedback: unresolved
Dispatch-->>Practitioner: Suggest alternative/combined technique
end
Estimated code review effort🎯 2 (Simple) | ⏱️ ~12 minutes Poem
Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✨ Finishing touches🧪 Generate unit tests (beta)
Comment |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Actionable comments posted: 0
🧹 Nitpick comments (4)
skills/architecture/ABOUT.md (1)
7-8: Wrap the repository URL to satisfy markdownlint.markdownlint (MD034) flags the bare URL on Line 7. Wrapping it in angle brackets resolves the warning while keeping the text identical.
-- URL: https://github.com/microsoft/amplifier +- URL: <https://github.com/microsoft/amplifier>skills/research/ABOUT.md (1)
7-8: Fix the bare URL to clear MD034.Line 7 still has an unwrapped URL, so markdownlint continues to warn. Angle brackets silence the rule without altering meaning.
-- URL: https://github.com/microsoft/amplifier +- URL: <https://github.com/microsoft/amplifier>skills/collaboration/executing-plans/SKILL.md (1)
72-78: Remove the duplicated blocker reminder.The new “Stop and Ask for Help” section already captures this guidance, yet Line 78 still repeats the old wording. Drop the duplicate to keep the Remember section focused.
- Reference skills when plan says to - Between batches: just report and wait -- Stop when blocked, don't guessskills/problem-solving/ABOUT.md (1)
7-8: Convert the bare URL to comply with MD034.Line 7 needs angle brackets so markdownlint stops flagging the link.
-- URL: https://github.com/microsoft/amplifier +- URL: <https://github.com/microsoft/amplifier>
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (13)
skills/architecture/ABOUT.md(1 hunks)skills/architecture/preserving-productive-tensions/SKILL.md(1 hunks)skills/collaboration/brainstorming/SKILL.md(3 hunks)skills/collaboration/executing-plans/SKILL.md(2 hunks)skills/problem-solving/ABOUT.md(1 hunks)skills/problem-solving/collision-zone-thinking/SKILL.md(1 hunks)skills/problem-solving/inversion-exercise/SKILL.md(1 hunks)skills/problem-solving/meta-pattern-recognition/SKILL.md(1 hunks)skills/problem-solving/scale-game/SKILL.md(1 hunks)skills/problem-solving/simplification-cascades/SKILL.md(1 hunks)skills/problem-solving/when-stuck/SKILL.md(1 hunks)skills/research/ABOUT.md(1 hunks)skills/research/tracing-knowledge-lineages/SKILL.md(1 hunks)
🧰 Additional context used
🪛 markdownlint-cli2 (0.18.1)
skills/architecture/ABOUT.md
7-7: Bare URL used
(MD034, no-bare-urls)
skills/research/ABOUT.md
7-7: Bare URL used
(MD034, no-bare-urls)
skills/problem-solving/ABOUT.md
7-7: Bare URL used
(MD034, no-bare-urls)
Updated plugin source format to use URL instead of GitHub repo because the github repo format breaks for users without github ssh keys configured. Fixes obra/superpowers#1
Documents all changes since v1.0.0: - Skills repository separation (breaking change) - New problem-solving and research skills (PR #1) - Personal superpowers system (PR #2, later replaced) - Auto-update functionality - Skills improvements (using-skills v4.0.0, sharing-skills v2.0.0) - Tools improvements (find-skills, skill-run) - Plugin infrastructure changes - Migration guide for existing users
Add problem-solving skills from amplifier patterns
Documents all changes since v1.0.0: - Skills repository separation (breaking change) - New problem-solving and research skills (PR obra#1) - Personal superpowers system (PR obra#2, later replaced) - Auto-update functionality - Skills improvements (using-skills v4.0.0, sharing-skills v2.0.0) - Tools improvements (find-skills, skill-run) - Plugin infrastructure changes - Migration guide for existing users
New epic (Superpowers-hfa) with 5 tasks: - TDD skill: revert discard→delete with Rule obra#1 compliance - writing-skills: add 'ask first' to No exceptions blocks - writing-skills: add test artifact cleanup step - testing-skills-with-subagents: add 'ask first' - Verification: grep checks for compliance Also reorganize plan files into bead_plans/ subdirectory. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Change all "Discard" to "Delete" (lines 37, 43, 244, 278, 301) - Remove stash escape hatch that defeated TDD's cognitive purpose - Add modified/new file distinction for Rule obra#1 compliance: - Modified files: git checkout (safe, no permission needed) - New files: ask permission before deleting - Update summary with modified/new clarification Closes Superpowers-n8s 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- writing-skills: add 'Ask your human partner before deleting' to Iron Law and Good example, update Red Flags summary - writing-skills: add test artifact cleanup step to Deployment checklist - testing-skills-with-subagents: add 'ask first' to After example Completes CLAUDE.md Rule obra#1 compliance for delete language. Closes Superpowers-hfa epic. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Adds 12 granular beads to track the removal of writing-plans and using-git-worktrees skills, structured for TDD execution: Phase 1 - Verification: - Superpowers-sht: Verify replacement skills exist Phase 2 - Pre-deletion updates: - Superpowers-hpm: Rename command write-plan → plan-to-beads - Superpowers-xz2: Update brainstorming skill - Superpowers-lts: Update finishing-a-development-branch skill - Superpowers-jlx/u70/ctj/9wt: Update README (4 changes) Phase 3 - Approval: - Superpowers-dnk: Human approval checkpoint (CLAUDE.md Rule obra#1) Phase 4 - Deletion (requires approval): - Superpowers-8wb: Delete skills/writing-plans/ - Superpowers-v5d: Delete skills/using-git-worktrees/ Phase 5 - Validation: - Superpowers-urp: Post-deletion validation and commit 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Summary
Extract and adapt problem-solving techniques from Amplifier's agent system into focused, composable skills.
New skills added:
Enhanced existing skills:
Key improvements:
What Changed
Problem-Solving Skills (New)
Architecture Skills (New)
Research Skills (New)
Workflow Skills (Enhanced)
Testing
All skills tested with RED-GREEN-REFACTOR methodology:
Attribution
Skills derived from amplifier agents (commit 2adb63f):
See ABOUT.md files in each category for full attribution.
Summary by CodeRabbit
New Features
Improvements
Documentation