65207a5200
feat(hooks): add plan-mode-game hook component
...
Adds a browser game (Dino Runner, Snake, Flappy Bird) that launches
automatically when Claude enters Plan Mode. Users can play while
Claude thinks, with a live activity panel showing tool calls in real time.
- EnterPlanMode / /plan → starts Node.js server on port 3456, opens browser
- Read/Bash/Edit/etc → streams tool activity to the activity panel
- Write to ~/.claude/plans/ → also triggers the game (for /plan command)
- AskUserQuestion → pauses the game until user responds
- ExitPlanMode → shows "Claude Termino!" overlay with confetti
Features: 3 games with high score persistence, signed score sharing URLs,
pause overlay when Claude needs input, activity feed with tool icons.
Requires Node.js and curl. macOS only (uses 'open' command).
Install globally in ~/.claude/settings.json to work across all projects.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-06-13 14:21:04 -04:00
9b1e38673a
Add flatten-mcp to devtools MCPs ( #632 )
2026-06-13 11:32:57 -04:00
56f2c921c7
feat(mcp): add memex temporal knowledge-graph MCP ( #609 )
...
Adds the memex MCP server under mcps/integration. memex watches git
repos and builds a temporal knowledge graph of modules, symbols,
decisions, and open problems (Graphiti + Neo4j), serving it to AI
coding agents over MCP. 12 read/write tools.
npx -y stifler-memex-mcp serve --repo .
2026-05-29 18:03:49 -04:00
d3b53e089b
Add pdf-fill-studio skill (local PDF form filling) ( #614 )
...
* Add pdf-fill-studio skill
* Use copy-paste-safe placeholders (no <...> shell redirection)
2026-05-29 17:51:08 -04:00
fca4c0e7e9
feat: add android-cicd skill for automated Google Play publishing ( #580 )
2026-05-23 15:53:13 -04:00
161b57d33d
Add bleu planning skill under development ( #598 )
2026-05-23 15:50:21 -04:00
a745deec0d
Add building-blog skill (web-development) ( #601 )
2026-05-23 15:49:00 -04:00
ac2e7790e6
fix: replace invalid git diff --since with correct two-step approach ( #602 )
2026-05-23 15:47:55 -04:00
258d901e49
fix: replace VS Code Copilot tool names with valid Claude Code tools in devops agents ( #603 )
2026-05-23 15:47:04 -04:00
2def1c02e4
fix: resolve duplicate skill names for -official document processing skills ( #604 )
2026-05-23 15:45:08 -04:00
7d6c136ad1
improve: enhance graphql-performance-optimizer based on automated review ( #596 )
...
- Add missing frontmatter fields: model, color, permissionMode
- Rewrite description with three structured examples (N+1 diagnosis, APQ+CDN caching, federated graph profiling)
- Replace deprecated graphql-depth-limit with @envelop/depth-limit
- Replace graphql-cost-analysis with graphql-query-complexity
- Add Persisted Queries / Trusted Documents section (APQ vs Trusted Documents decision table)
- Add Federation Performance section (router query plan caching, subgraph DataLoader scoping, __resolveReference batching)
- Add Subscription Scaling section (graphql-ws, Redis PubSub, SSE via graphql-sse, withFilter)
- Replace naive getRequestedFields helper with graphql-parse-resolve-info recommendation
- Add note to defer security content to graphql-security-specialist agent
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-18 09:10:16 -04:00
eca0780250
chore: Sync Anthropic skills upstream changes
...
- claude-api: Update model config shorthand (drop `type` field from example)
- slack-gif-creator: Update LICENSE copyright to Anthropic, PBC 2026
2026-05-18 08:38:37 -04:00
2e6222897d
Add Explorium MCP to web-data ( #587 )
...
* Add Explorium MCP to web-data
* Add api_key header for API key auth; OAuth supported by removing headers
2026-05-15 20:41:05 -04:00
b37291c508
Add career skills category (20 skills for job-search workflow) ( #590 )
...
Introduces a new `career/` skill category under
cli-tool/components/skills/ with 20 skills covering resumes, cover
letters, LinkedIn, interview prep, salary negotiation, and offer
comparison. Each skill follows the SKILL.md format with kebab-case
name matching its directory and a trigger-rich description.
Skills included:
- Resume writing: resume-tailor, resume-bullet-writer,
resume-quantifier, resume-formatter, resume-section-builder,
resume-version-manager, resume-ats-optimizer
- Role-specific: tech-resume-optimizer, executive-resume-writer,
creative-portfolio-resume, academic-cv-builder
- Application: cover-letter-generator, linkedin-profile-optimizer,
reference-list-builder, portfolio-case-study-writer
- Interview/negotiation: job-description-analyzer,
interview-prep-generator, salary-negotiation-prep,
offer-comparison-analyzer
- Career pivots: career-changer-translator
Source plugin: https://github.com/m0k0ut/resume-skills (MIT).
2026-05-15 20:33:35 -04:00
a8bac27432
feat: add luma-imagegen skill for Luma AI image generation
...
Adds a new skill in creative-design/luma-imagegen that generates images
via the Luma AI Dream Machine API (Photon models). Includes interactive
prompt collection, API key detection from env/.env, async polling, and
image download. Uses only Python stdlib — no extra dependencies.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-15 19:23:24 -04:00
0dd8b11d48
improve: enhance blockchain-developer based on automated review ( #588 )
...
- Replace hardcoded progress metrics with actual-session placeholders
- Add model: sonnet and color: yellow to frontmatter
- Remove fictional Blockchain Context Assessment JSON protocol
- Add human-in-the-loop pause criteria section (When to Stop and Ask)
- Modernize security toolchain: Slither + Echidna/Foundry fuzzing + Certora/SMTChecker
- Add missing 2025-critical EIPs: ERC-4337, EIP-7702, EIP-7201, EIP-1153
- Replace flat bullet inventory with behavior-driving instructions (gas, security, testing)
- Add WebSearch and WebFetch to tools list
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-14 17:10:02 -04:00
eeb652bc4e
improve: enhance business-analyst based on automated review ( #584 )
...
- Replace discovery step 1 with concrete user-facing questions (remove non-existent context-manager reference)
- Add Human-in-the-Loop pause criteria section with 5 explicit stop conditions
- Replace hardcoded fictional metrics (87, $2.3M, 23 stakeholders) with dynamic placeholder tokens and explicit instruction to use only session-derived data
- Collapse 9 redundant bullet-list sections into focused behavioral directives
- Add opinionated process modeling defaults (BPMN 2.0, value stream mapping, current-state-first)
- Remove Edit tool (analysis-focused agent; Write sufficient for BRD/spec authoring)
- Add model: sonnet to frontmatter
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-13 17:10:20 -04:00
82b0fb3366
feat: migrate 13 Google Cloud skills from google/skills ( #583 )
...
* feat: migrate 13 Google Cloud skills from google/skills repository
Adds Apache 2.0-licensed skills from https://github.com/google/skills
covering Google Cloud platform services and well-architected practices.
Skills added:
- ai-research/gemini-api-agent-platform
- database/alloydb-basics, bigquery-basics, cloud-sql-basics
- development/cloud-run-basics, firebase-basics, gke-basics
- development/google-cloud-onboarding, google-cloud-networking-observability
- development/google-cloud-waf-reliability, google-cloud-waf-cost-optimization
- security/google-cloud-auth, google-cloud-waf-security
Each skill preserves upstream content and supporting reference docs, with
a source: google/skills (Apache 2.0) frontmatter line for attribution.
Catalog regenerated and copied to dashboard/public/components.json.
https://claude.ai/code/session_01TzH6kveVv8jxFjg8oMMtkZ
* chore: refresh component catalog download stats
Catalog regenerated after Google skills migration picks up updated
download counts from Supabase.
https://claude.ai/code/session_01TzH6kveVv8jxFjg8oMMtkZ
---------
Co-authored-by: Claude <noreply@anthropic.com >
2026-05-13 09:53:18 -04:00
cf931873b9
improve: enhance graphql-architect based on automated review ( #581 )
...
- Update Apollo Federation version reference from 2.5 to 2.10+ with @link directive and federated subscriptions notes
- Add model, color, and permissionMode frontmatter fields
- Narrow tools list to Read, Grep, Glob, Edit, Write (remove Bash)
- Replace schema-first checklist item with SDL-first vs code-first section (Pothos, TypeGraphQL, graphql-codegen)
- Add GraphQL Yoga and Envelop as server alternatives with selection criteria
- Remove vestigial JSON context manager protocol block; replace with prose discovery instructions
- Replace abstract testing checklist with concrete named tooling (graphql-inspector, graphql-ws, autocannon, etc.)
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-12 17:09:55 -04:00
9cef8df9c1
improve: enhance api-designer based on automated review
...
- Replace fictional context-manager JSON blocks with real tool-based discovery (Glob/Grep) in When Invoked section
- Add OpenAPI 3.1 YAML snippet with security scheme, resource schema, error schema, and pagination
- Add GraphQL SDL example with Relay connection-based pagination and @deprecated directive
- Add Deliverables section specifying openapi.yaml, schema.graphql, MIGRATION.md, API-DECISION.md outputs
- Add model: sonnet and color: cyan frontmatter fields
- Expand description with gRPC/protocol-selection and fourth microservices example
- Add protocol selection comparison table (REST vs GraphQL vs gRPC)
- Update OAuth 2.0 to OAuth 2.1 with PKCE enforcement, remove implicit flow
- Add Bash usage constraint limiting shell to API linters only
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-12 11:16:27 -04:00
90489f6212
improve: enhance database-architect based on automated review ( #574 )
...
- Add rich description with 3 example blocks (greenfield schema, tech selection, data migration)
- Add model: sonnet and Glob, Grep to frontmatter tools
- Remove hardcoded credential from read replica example; use ${REPLICATION_PASSWORD}
- Add structured "When Invoked" workflow section (4 steps)
- Expand technology selection matrix with vector, graph, serverless_relational categories
- Add multi-tenant and RLS patterns section with isolation strategy comparison table
- Add agent integration section listing collaboration boundaries with 5 agents
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-09 19:10:14 -04:00
a9c36fc852
feat: add 3 subagent hooks components (security theme)
...
- hooks/security/ai-bash-guard: PreToolUse with type:agent — AI-powered bash security guard using haiku subagent
- settings/hooks/subagent-lifecycle-logger: SubagentStart/SubagentStop audit trail to .claude/agent-log.jsonl
- agents/security/read-only-auditor: security auditor with frontmatter hooks blocking Write/Edit/Bash while active
2026-05-08 20:27:16 -04:00
004cf8460b
improve: enhance task-decomposition-expert based on automated review ( #573 )
...
- Rewrote description with 3 concrete <example> blocks and <commentary> invocation triggers
- Added model: sonnet field
- Expanded tools to Read, Write, Edit, Bash, Glob, Grep, WebSearch
- Removed mandatory ChromaDB coupling from critical path entirely
- Added Requirements Gathering as mandatory first step with 6 intake questions
- Added concrete output format: WBS with 8/80 rule, dependency graph notation, parallelism map table, risk register table, validation checkpoints
- Added explicit Agent Handoff Plan section mapping workstreams to specialist agents
- Added Integration with Other Agents section
- Added structured JSON progress format and completion message template
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-08 17:10:45 -04:00
a0dfc1bb18
improve: enhance moc-agent with spec-compliant format and relative paths ( #572 )
...
- Remove all hardcoded absolute paths (/Users/cam/VAULT01/...) and replace with relative paths rooted at ./
- Replace VAULT01-specific references with generic vault terminology
- Add model: sonnet field to frontmatter
- Add Grep and Edit to tools list; remove redundant LS
- Expand description with 3 concrete <example>/<commentary> blocks
- Add Native Fallback Workflow using Glob+Grep (no external script dependency)
- Make Script-Assisted Workflow optional with clear instructions to run from vault root
- Add MOC Hierarchy section explaining LYT three-tier system (Home/Topic/Sub-MOC)
- Add optional Dataview block to MOC template with note to remove if plugin absent
- Add vault-structure validation guidance before writing MOCs
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-07 17:10:39 -04:00
04ec8c3d8b
sync: update skill-creator from anthropics/skills
...
Major upstream rewrite with eval-driven iterative workflow:
- New agents/ dir (analyzer, comparator, grader)
- New eval-viewer/ for benchmark review UI
- New scripts for eval runs, benchmarking, and description optimization
- Updated SKILL.md with new process-oriented approach
- Preserved local-only files (output-patterns.md, workflows.md, init_skill.py)
2026-05-07 11:07:15 -04:00
74d4b1036b
improve: enhance search-specialist with spec-compliant format and current tooling ( #569 )
...
- Add tools and model frontmatter fields (WebSearch, WebFetch; sonnet)
- Rewrite description with three concrete examples using example/commentary tags
- Add imperative "When Invoked" seven-step workflow checklist
- Add iterative retrieval loop section with explicit stopping conditions
- Add source credibility evaluation framework as a scored table
- Expand contradiction handling into an explicit protocol with example format
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-06 17:09:58 -04:00
963b44fe22
sync: update claude-api skill from anthropics/skills
...
Add Managed Agents outcomes, multiagent, and webhooks docs.
Update API reference, events, core, overview, tools, onboarding,
and live-sources with latest upstream changes.
2026-05-06 13:24:23 -04:00
e9232563c4
improve: enhance model-evaluator with spec-compliant format and current tooling ( #566 )
...
- Rewrote description with three <example> blocks covering task-specific model selection, benchmark design, and post-deployment regression testing
- Added Standard Frameworks & Tools section with HELM, lm-evaluation-harness, DeepEval, RAGAS, Promptfoo, and Chatbot Arena
- Updated Model Categories to tier-based language (Haiku/Sonnet/Opus) and current model names (GPT-4o, Gemini 1.5/2.0), removed deprecated version numbers
- Added model: sonnet frontmatter and expanded tools list (added Edit, Glob, Grep)
- Replaced skeletal Python stub with Statistical Requirements section (sample sizes, CI, effect size, Cohen's kappa, Bonferroni correction)
- Added Integration with Other Agents section mapping handoffs to llm-architect, prompt-engineer, and ai-ethics-advisor
- Added Step 6 Post-Deployment Monitoring (drift detection, re-evaluation triggers, Arize Phoenix, LangSmith, Promptfoo CI)
- Removed emoji from output template header
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-05 17:10:17 -04:00
f6b5a370b8
Add diagrammer skill ( #564 )
2026-05-04 19:15:53 -04:00
43742e2ff3
improve: enhance llms-maintainer with spec-compliant format and framework detection ( #563 )
...
- Fix llms.txt skeleton to use official Markdown spec (H1, blockquote, H2 sections, link entries)
- Fix page entries to use standard `- [Title](url): description` format instead of invented key:value
- Add framework detection step (Astro, Nuxt, Next.js, Svelte, Hugo) to determine output path
- Add `model: haiku` to frontmatter for cost-efficient deterministic scanning
- Add `maxTurns: 20` to prevent runaway iteration on large sites
- Remove unconditional `git push` — commit only by default, user pushes manually
- Expand "AEO" to "AI Engine Optimization (AEO)" on first use in description and closing line
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-04 17:10:11 -04:00
20f38d42fb
chore: sync Anthropic skills upstream changes
...
- Update LICENSE.txt copyright to "Copyright 2026 Anthropic, PBC." across 10 skills
- Remove deprecated `purpose` parameter from managed agents file upload examples in claude-api skill
Source: https://github.com/anthropics/skills
2026-05-04 11:07:12 -04:00
d9e3036e66
improve: enhance api-architect with GraphQL, security, and scoped tooling ( #561 )
...
- Rewrote description with 3 trigger examples using <example>/<commentary> tags
- Expanded scope to cover full GraphQL architecture: SDL-first vs code-first,
resolver pattern, DataLoader for N+1 prevention, federation, persisted queries,
query depth/complexity limiting, and production introspection disabling
- Removed erroneous "Code Interpreter" reference; replaced with Write/Edit tool instruction
- Added mandatory Security Checklist section covering TLS, auth schemes, rate limiting,
OWASP API Security Top 10, and GraphQL-specific controls
- Removed Bash from tools (Read, Grep, Glob, Edit, Write only); added permissionMode: acceptEdits
- Added API versioning and lifecycle guidance (URL/header/query-param for REST,
@deprecated directive lifecycle for GraphQL)
- Added model: sonnet and color: blue frontmatter fields
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-03 17:10:21 -04:00
f75f4e8add
improve(javascript-pro): modernize runtime, tooling, and framework coverage ( #560 )
...
- Update persona from Node.js 20+ to Node.js 22 LTS / Node.js 24 LTS
- Expand ES mastery from ES2023 to ES2025 (Object.groupBy, Promise.withResolvers, Set methods, Iterator helpers, RegExp.escape, explicit resource management)
- Add Runtime ecosystem section covering Node.js 22/24, Bun, and Deno 2.x
- Replace Jest-first testing with Vitest as default; add Playwright for E2E and node:test for library authors
- Modernize Build and tooling: Vite 6/7/8 (Rolldown-powered) as default, Turbopack, Rolldown, esbuild; demote Webpack to legacy
- Add Modern framework patterns section: React 19, Next.js 15, Vue 3.5/3.6, Svelte 5, SolidJS
- Add Package management section: npm workspaces, pnpm, Bun, exports field, provenance
- Expand Security practices with Socket.dev/Snyk, ignore-scripts, lockfile CI verification, typosquatting awareness
- Update third description example to reference Vitest instead of Jest
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-02 17:09:54 -04:00
edc8f2edf2
improve(research-synthesizer): enhance with phased workflow, tool expansion, and quality verification ( #559 )
...
- Add WebSearch and WebFetch to tools with sparingly-use guidance
- Add Input Discovery Protocol for locating upstream researcher outputs and handling missing sources gracefully
- Replace flat bullet lists with 4-phase execution framework (Discovery, Extraction, Integration, Output)
- Add Quality Verification Checklist before final output
- Require standalone synthesis-summary.md written before synthesis.json
- Add missing_researchers field to synthesis_metadata schema
- Add single_source consensus_level value
- Replace abstract examples with realistic Claude Code workflow scenarios referencing actual agent outputs
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-05-01 17:10:15 -04:00
6db58d321b
improve(hackathon-ai-strategist): enhance with context gathering, phased execution, and pitch structure ( #558 )
...
- Rewrote description with three trigger-oriented examples (pre-ideation, mid-triage, pitch prep)
- Added model: sonnet to frontmatter
- Added Required Initial Step: Context Gathering section (6 fields, blocks advice until complete)
- Added Time-Boxed Execution Framework with 5 phases, explicit go/no-go checkpoints for 24h hackathon
- Added Sponsor Strategy and Prize-Track Optimization section with scoring decision framework
- Added 3-Minute Pitch Outline with time-annotated segments and Demo Reliability Checklist
- Changed 'fake for demos' to 'stub or mock for the demo' to remove ethical ambiguity
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-04-30 17:10:26 -04:00
f27640039e
improve(llm-architect): enhance with structured decision guidance and evaluation framework ( #556 )
...
- Replace flat bullet-point taxonomy with hierarchical H2/H3 decision trees for serving, fine-tuning, RAG, and safety
- Add Required Initial Step section for mandatory requirements gathering (latency, throughput, model class, infra, compliance)
- Add Evaluation and Observability section with RAGAS v0.4+ metrics, LLM-as-judge guidelines, and observability stack
- Add model: sonnet to frontmatter for complex system design reasoning
- Replace hardcoded fictional metrics (187ms, 127 tokens/s) with placeholder templates
- Update tools list: add WebSearch, remove Glob and Grep
- Update assistant examples in description to reflect requirements-first approach
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-04-29 23:27:32 -04:00
8b18c46f67
Add trace-mcp MCP server to devtools ( #553 )
...
trace-mcp is a framework-aware code intelligence MCP server providing
semantic navigation, impact analysis, and refactoring tools across
60 framework integrations and 81 languages.
Repo: https://github.com/nikolai-vysotskyi/trace-mcp
npm: https://www.npmjs.com/package/trace-mcp
2026-04-29 14:45:21 -04:00
262da50f98
fix(context-timeline): detect Opus 1M context window dynamically ( #548 )
...
CONTEXT_LIMIT was hardcoded at 200_000 tokens. Users on Claude Opus 4.7
with the 1M context window (model: "opus[1m]") saw the dashboard report
80–90% usage when actual usage was 16–18%, producing constant false
alarms and making the meter unusable on long sessions.
Replace the constant with _detect_context_limit(cwd) which resolves the
limit in this order:
1. CONTEXT_TIMELINE_LIMIT env var (explicit override).
2. "[1m]" / "[1000k]" / "[200k]" suffix in the model field of
<cwd>/.claude/settings.json or ~/.claude/settings.json.
3. 200_000 fallback (current behaviour).
The limit is resolved once per _Builder instance, not per update tick,
so there is no per-event filesystem cost.
No new dependencies (still stdlib only). Default behaviour is unchanged
for users without [1m] in settings.
Verified live on a real Opus 1M session of ~178K tokens: meter went
from 89% (red, false alarm) to 17.9% (cyan, correct).
2026-04-27 20:10:27 -04:00
e11ff2e105
fix(context-timeline): cross-platform path encoding for transcript lookup ( #550 )
...
The `_encode_cwd` function only replaced `/` with `-`, which fails on
Windows where paths use `\` and include drive letters like `D:`.
Use regex to replace any non-alphanumeric character with `-`.
This handles all edge cases across platforms in a single expression.
Before: D:\Claude\mcx -> lookup failed
After: D:\Claude\mcx -> D--Claude-mcx (matches Claude Code encoding)
2026-04-27 20:09:13 -04:00
e44ddb78ca
improve: enhance accessibility-tester agent based on automated review ( #546 )
...
- Add correct YAML frontmatter with name, description, tools, and model fields
- Target WCAG 2.2 explicitly with all 9 new criteria listed in reference table
- Mandate hybrid testing methodology (automated + manual) with 40% coverage note
- Restrict tools to Read, Grep, Glob, Bash — no Write/Edit (auditor pattern)
- Set model to sonnet for deep ARIA and compliance reasoning
- Add two example blocks with user/assistant/commentary tags
- Add legal compliance mapping to ADA, Section 508, and EAA (June 2025)
- Add structured finding format with WCAG criterion, severity, remediation, verification
- Add summary scorecard template with severity counts and WCAG 2.2 criterion status
- Add ARIA patterns reference (dialog, combobox, tabs, navigation, live regions)
- Add screen reader test matrix (VoiceOver, NVDA, TalkBack, JAWS)
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-04-26 17:10:24 -04:00
fc2155655c
fix(context-timeline): reload sessions from disk on cache miss when daemon predates session
2026-04-25 23:24:30 -04:00
99f423049d
feat(hook): add monitoring/context-timeline real-time session visualizer
...
Persistent local HTTP daemon (port 7878, 1h watchdog) that reads Claude Code
JSONL transcripts and renders a git-graph timeline in the browser — one column
per agent (main + subagents), tool nodes color-coded by type, context-window
sidebar with cache/input/output breakdown. Pure stdlib Python, zero pip deps.
2026-04-25 23:08:28 -04:00
dc8d99bbc5
improve: enhance ai-ethics-advisor based on automated review ( #542 )
...
- Rewrote description with concrete usage triggers and 3 inline examples
- Added ISO/IEC 42001 (AI Management System) and ISO/IEC 42005 (AI System Impact Assessment) to Regulatory Compliance Framework
- Added UNESCO AI Ethics Recommendation (2021) with all 10 principles
- Added new "Agentic AI Ethics" section covering goal manipulation, minimal footprint, oversight checkpoints, inter-agent trust, and tool misuse surface
- Expanded tools list to include Edit, Bash, and Glob
- Replaced broken BiasMonitor Python skeleton with production-ready tooling inventory (IBM AIF360, Fairlearn, What-If Tool, Alibi Detect)
- Added "Output Artifacts" subsection defining 5 concrete deliverable files
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-04-25 22:10:19 -04:00
5948b4716f
improve: enhance data-scientist agent based on automated review ( #541 )
...
- Add model field (claude-sonnet-4-5) to frontmatter
- Remove duplicate Statistical methods and ML algorithms sections
- Replace non-functional context manager JSON block with plain-language context gathering instructions
- Modernize Tools & libraries: add Polars, DuckDB, MLflow, Great Expectations/Pandera, CatBoost, Altair
- Add Responsible Analysis section covering bias auditing, data privacy, reproducibility, transparency, and fairness metrics
- Add reproducibility and fairness items to data science checklist
- Remove hardcoded fabricated metrics from progress tracking JSON block
- Merge ANOVA/MANOVA into Statistical modeling section from removed duplicate
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-04-25 17:10:33 -04:00
0eec42a824
chore: sync claude-api skill from anthropics/skills (April 2026)
...
Adds memory_stores support to Managed Agents: new managed-agents-memory.md
reference, updated beta headers, reading guide, and API reference docs.
2026-04-24 15:27:14 -04:00
b066c26983
chore: sync pdf skill from anthropics/skills (April 2026)
...
Updated SKILL.md (new subscripts/superscripts section, improved description),
forms.md (simplified structure extraction workflow), and 7 scripts.
Added new extract_form_structure.py script.
2026-04-20 11:20:35 -04:00
4439f9b3f2
improve: enhance debugger agent based on automated review ( #531 )
...
- Add model: claude-sonnet-4-5 frontmatter field for consistent reasoning quality
- Remove fictional JSON Communication Protocol and Progress Tracking blocks
- Replace generic 3-phase workflow with 6-step fault-localization decision tree
- Add Observability-Driven Debugging section (traces, logs, change correlation)
- Trim description from ~2300 to ~1200 chars by tightening commentary blocks
- Remove non-executable rubber duck debugging technique
Automated review cycle | Co-Authored-By: Claude Code <noreply@anthropic.com >
2026-04-18 20:06:30 -04:00
5d84560b69
chore: sync claude-api skill from anthropics/skills (April 2026) ( #530 )
...
Synced cli-tool/components/skills/development/claude-api/ with upstream
anthropics/skills repo (commits March 22 – April 16, 2026).
Updated existing files (21):
- SKILL.md: grew from 20,548 to 32,987 bytes (Managed Agents dispatch logic)
- curl/examples.md, java/claude-api.md, php/claude-api.md, ruby/claude-api.md
- python/claude-api/{README,batches,files-api,streaming,tool-use}.md
- typescript/claude-api/{README,batches,files-api,streaming,tool-use}.md
- shared/{error-codes,live-sources,models,prompt-caching,tool-use-concepts}.md
- shared/managed-agents-{api-reference,client-patterns,core,environments,
events,onboarding,overview,tools}.md (previously existed, now updated)
- curl/managed-agents.md (previously existed, now updated)
Added new files (7 language + 1 shared):
- curl/managed-agents.md (new entry; prior file was empty placeholder)
- go/managed-agents/README.md
- java/managed-agents/README.md
- php/managed-agents/README.md
- python/managed-agents/README.md
- ruby/managed-agents/README.md
- typescript/managed-agents/README.md
- shared/agent-design.md
- shared/model-migration.md (65 KB consolidated migration guide)
Local-only files preserved (not touched):
- python/agent-sdk/{README,patterns}.md
- typescript/agent-sdk/{README,patterns}.md
Also updated:
- cli-tool/components/skills/ANTHROPIC_ATTRIBUTION.md: Last Updated → April 18, 2026
- Regenerated docs/components.json and dashboard/public/components.json
2026-04-18 19:01:17 -04:00
bb5fa62226
Merge pull request #519 from socialproKGCMG/add-szamlazz-command
...
Add szamlazz Hungarian invoicing command
2026-04-18 18:37:13 -04:00
cbaaec3257
Merge pull request #520 from yurukusa/add-shell-wrapper-guard
...
Add shell-wrapper-guard: detect destructive commands in interpreter one-liners
2026-04-18 18:36:20 -04:00
daniel