feat(skills): SKL-REMED — full library remediation + 11 new skills (Sprints 0-8)#35
Open
WhatInTheShell wants to merge 3 commits into
Open
feat(skills): SKL-REMED — full library remediation + 11 new skills (Sprints 0-8)#35WhatInTheShell wants to merge 3 commits into
WhatInTheShell wants to merge 3 commits into
Conversation
Adds 6 NIST 800-61r3 IR analysis skills (after-action reviewer, CSF mapper, gap analyzer, maturity scorer, policy reviewer, recommendation auditor), accompanying agent definition, and project-manager OS dashboard skill. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Sprint 0 — P0 bug fixes: - domain-name-brainstormer: remove fabricated availability claims - jira: fix MCP namespace (mcp__atlassian__ → mcp__claude_ai_Atlassian_Rovo__), remove broken jira-link-issues ref - mui: fix Grid v2 size prop syntax, replace broken useMuiSnackbar hook ref - meme-factory: resolve ~n vs %0A encoding contradiction (canonical: %0A) - reducing-entropy: inline worse-is-better philosophy, fix broken reference - professional-communication: wire orphaned remote-async-communication.md trigger Sprint 1 — Library templates: - NEVER-list-template.md, Mindset-Navigation-Philosophy-template.md - references-directory-convention.md, README-policy.md, error-handling-stub-template.md Sprints 2-5 — All 50 skills remediated: - Canonical Mindset/Navigation/Philosophy/NEVER/When-Things-Go-Wrong sections added to every skill - 19 F-tier skills: full structural rebuilds with expert knowledge injection - 16 D-tier skills: NEVER lists expanded, error handling added, references/ created - 15 C-tier skills: targeted gap fixes, inline content moved to references/ Sprint 7 — Structural repairs + new skills: - New skills: skill-creator, api-design, security-review, nist-800-61r3-incident-coordinator - NIST CSF element registry consolidated into nist-800-61r3-shared/references/ - react-dev: 461 → 96 lines; 5 reference files created - 10 cross-skill Navigation callouts wired - 5 frontmatter descriptions improved (react-dev, humanizer, command-creator, game-changing-features, ship-learn-next) - Orphaned root-level files relocated to references/ (react-useeffect, writing-clearly-and-concisely) - session-handoff evals/ directory removed (Pattern 8 auxiliary files) Library baseline → post-remediation: - Average score: 75/120 (D) → est. 87-90/120 (C+) - F-grade skills: 19 → 0 - Total skills: 50 → 55 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…ons merge New skills: - git-workflow: branch strategy, rebase/merge decision, bisect mastery, stash hygiene, worktree patterns, reflog recovery (77L SKILL.md + 5 references) - code-review: severity calibration, diff vs file review, 3-category feedback taxonomy, language anti-patterns, test review protocol (165L + 3 references) - docker-and-containers: multi-stage build patterns, layer cache invalidation, image size, health checks, compose startup ordering, container security (126L + 2 references) - testing-strategy: pyramid vs trophy, test double taxonomy, Playwright vs Cypress, flakiness root causes, mutation testing, boundary patterns (161L + 3 references) - presentation-to-audience: narrative arc selection, audience calibration, opening strategy, Q&A bridge technique, cognitive load design (122L + 2 references) - monorepo-workspace: Nx/Turborepo selection, build graph, shared config inheritance, circular dependency detection, publishing strategy, remote caching (3 references) Merge: - feedback-mastery + difficult-workplace-conversations → workplace-conversations - Both original skills now have redirect stubs pointing to merged skill Library: 55 → 61 skills total Co-Authored-By: Claude Sonnet 4.6 <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
2 participants
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
Full library-wide skill quality remediation across 8 sprints, driven by the skill-judge rubric (120-point evaluation across 8 dimensions).
Baseline → After:
What Changed
Sprint 0 — P0 Bug Fixes
domain-name-brainstormer: removed fabricated domain availability claims (Claude has no WHOIS access)jira: fixed all MCP tool namespaces (mcp__atlassian__*→mcp__claude_ai_Atlassian_Rovo__*); removed non-existentjira-link-issuesscript refmui: fixed Grid v2sizeprop syntax; replaced brokenuseMuiSnackbarwith standard patternsmeme-factory: resolved~nvs%0Aencoding contradiction (canonical:%0Aper memegen.link API)reducing-entropy: inlined worse-is-better philosophy; fixed broken referenceprofessional-communication: wired orphanedremote-async-communication.mdloading triggerSprint 1 — Library Templates
4 library-wide convention files created at the root:
NEVER-list-template.md— canonical standard: 5+ items with non-obvious WHY requiredMindset-Navigation-Philosophy-template.md— canonical header blockreferences-directory-convention.md+README-policy.mderror-handling-stub-template.mdSprints 2–5 — All 50 Skills Remediated
Every skill received canonical
## Mindset / ## Navigation / ## Philosophy / ## NEVER / ## When Things Go Wrongsections. Heavy inline content moved toreferences/subdirectories.Sprint 7 — Structural Repairs
react-dev: 461 → 96 lines; 5 reference files extractednist-800-61r3-shared/references/csf-element-registry.mdreferences/;session-handoff/evals/removedSprint 7 — New Skills
skill-creator— authoring new skills from scratch; description-first methodologyapi-design— cursor pagination, HTTP 422 vs 400, PATCH absent≠null, OAuth2 flow selectionsecurity-review— React XSS surfaces, ORM injection vectors, JWT validation, TypeScript runtime boundarynist-800-61r3-incident-coordinator— live incident coordination completing the NIST clusterSprint 8 — New Skills + Merge
git-workflow— branch strategy, bisect mastery, stash hygiene, worktree patterns, reflog recoverycode-review— severity calibration, diff vs file review, 3-category feedback taxonomydocker-and-containers— multi-stage builds, layer cache, signal handling, container securitytesting-strategy— test double taxonomy, Playwright vs Cypress, flakiness root causes, boundary patternspresentation-to-audience— narrative arc selection, audience calibration, Q&A bridge techniquemonorepo-workspace— Nx/Turborepo selection, build graph, circular dependency detection, publishing strategyfeedback-mastery+difficult-workplace-conversations→workplace-conversations(near-identical scope; both now have redirect stubs)Test plan
SKILL.mdfiles have valid YAML frontmatterjiraMCP tool names resolve in active sessions (mcp__claude_ai_Atlassian_Rovo__*)domain-name-brainstormerno longer claims to verify availabilityreact-devreferences/ loading triggers work as expectedfeedback-masteryanddifficult-workplace-conversationsredirect stubs point toworkplace-conversations🤖 Generated with Claude Code