foryourhealth111-pixel/deepagent-toolchain-plan
bundled/skills/deepagent-toolchain-plan/SKILL.md
DeepAgent-style tool discovery for VCO: propose a minimal skill/tool chain (with verification points) and reduce confirm_required friction.
$ install --global
skillsauthnpx skillsauth add foryourhealth111-pixel/vco-skills-codex deepagent-toolchain-planInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 4, 2026, 11:05 PM176.0s1 file scanned
SKILL.md
- name:
- deepagent-toolchain-plan
- description:
- DeepAgent-style tool discovery for VCO: propose a minimal skill/tool chain (with verification points) and reduce confirm_required friction.
DeepAgent Toolchain Plan (VCO)
When to use
Use this skill when:
- VCO router returns
route_mode=confirm_requiredand you want a better, evidence-backed choice - The task spans multiple domains/tools and you need a skill chain, not a single skill
- The user asks “用什么工具/技能最好？” / “怎么编排这套技能？”
- The conversation is long or messy and you need to re-anchor on goal → toolchain → verification
Non-goals (avoid redundancy)
- This is not a replacement for VCO routing. It is an augmentation that proposes a chain.
- This is not GitNexus. For code dependency/impact, use GitNexus overlays.
- This does not introduce long-term episodic memory (VCO governance disables it).
Runtime (Upstream vendoring)
DeepAgent upstream is vendored for reference / optional advanced runs:
C:\Users\羽裳\.codex\_external\ruc-nlpir\DeepAgent\
VCO-managed runtime config and self-check scripts (no secrets stored/printed):
C:\Users\羽裳\.codex\skills\vibe\config\ruc-nlpir-runtime.jsonpwsh C:\Users\羽裳\.codex\skills\vibe\scripts\ruc-nlpir\preflight.ps1
Core output (must)
Return a toolchain with:
- Goal + deliverable (1–2 lines)
- Chain steps (3–8 steps, each: skill/tool + why + expected artifact)
- Verification points (at least 1 falsifiable check)
- Fallbacks (what to do if a tool is unavailable)
Workflow
Step 1: Capture the task in a contract
- Goal (one sentence)
- Deliverable (code / plan / report / dataset / etc.)
- Constraints (time, no heavy deps, offline-only, etc.)
Step 2: Ask VCO router for a white-box view (recommended)
Run the router script in probe mode to get candidates + overlays in a machine-readable form:
pwsh C:\Users\羽裳\.codex\skills\vibe\scripts\router\resolve-pack-route.ps1 -Prompt "<PROMPT>" -Grade L -TaskType planning -Probe -ProbeLabel "toolchain" -ProbeOutputDir outputs/runtime/router-probes
Then use the emitted confirm_ui + overlay advice to decide the chain.
Step 3: Build a minimal chain (DeepAgent principle)
Prefer a chain that:
- Starts with evidence acquisition (local docs / web / code graph)
- Then planning
- Then execution
- Ends with verification + review
Step 4: Guardrails
- If the chain requires web browsing, explicitly choose between:
web.run(fast structured browse)playwright/turix-cua(dynamic/interactive)
- If the chain requires heavy model hosting (vLLM), provide a Lite alternative.
Suggested chains (templates)
A) “Research → report”
webthinker-deep-research(Lite) →outputs/webthinker/.../report.mdflashrag-evidence(local protocol checks) → citeable snippetscode-reviewer(if code changes) orverification-quality-assurance(if routing changes)
B) “VCO enhancement work (config/skills)”
flashrag-evidence(locate existing policy/overlays)writing-plans(implementation plan with file paths + verify steps)verification-before-completion(run check + router probe)
Related Skills
foryourhealth111-pixel/shap
development
VerifiedTrustedCommunity
Model interpretability and explainability using SHAP (SHapley Additive exPlanations). Use this skill when explaining machine learning model predictions, computing feature importance, generating SHAP plots (waterfall, beeswarm, bar, scatter, force, heatmap), debugging models, analyzing model bias or fairness, comparing models, or implementing explainable AI. Works with tree-based models (XGBoost, LightGBM, Random Forest), deep learning (TensorFlow, PyTorch), linear models, and any black-box model.
1,760SKILL.mdUpdated Apr 27, 2026
foryourhealth111-pixel/sentry
development
VerifiedTrustedCommunity
Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry API; perform read-only queries with the bundled script and require `SENTRY_AUTH_TOKEN`.
1,760SKILL.mdUpdated Apr 27, 2026
foryourhealth111-pixel/senior-prompt-engineer
development
VerifiedTrustedCommunity
World-class prompt engineering skill for LLM optimization, prompt patterns, structured outputs, and AI product development. Expertise in Claude, GPT-4, prompt design patterns, few-shot learning, chain-of-thought, and AI evaluation. Includes RAG optimization, agent design, and LLM system architecture. Use when building AI products, optimizing LLM performance, designing agentic systems, or implementing advanced prompting techniques.
1,760SKILL.mdUpdated Apr 27, 2026
foryourhealth111-pixel/senior-ml-engineer
development
VerifiedTrustedCommunity
World-class ML engineering skill for productionizing ML models, MLOps, and building scalable ML systems. Expertise in PyTorch, TensorFlow, model deployment, feature stores, model monitoring, and ML infrastructure. Includes LLM integration, fine-tuning, RAG systems, and agentic AI. Use when deploying ML models, building ML platforms, implementing MLOps, or integrating LLMs into production systems.
1,760SKILL.mdUpdated Apr 27, 2026