wkharness-guards
Provides management and unblocking paths for cluster security guards in wkharness environments.
Install
mkdir -p .claude/skills/wkharness-guards && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/18690" && unzip -o skill.zip -d .claude/skills/wkharness-guards && rm skill.zipInstalls to .claude/skills/wkharness-guards
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Guard cluster quick-ref for wkharness.ps1. One-liner per guard + escape hatch. Details in howto and ref skills.Key capabilities
- →Reference guard cluster G1-G6 for `wkharness.ps1`
- →Reference guard cluster G7-G12 for `wkharness.ps1`
- →Understand Agent guards like SPEC-GATE and CLAUDE-MD-SYNC
- →Utilize deadlock escape procedures
- →Understand guard cascade order
- →Clear `SKILL-UPDATE-PENDING` lock
How it works
The skill provides a quick reference for `wkharness.ps1` guard clusters, Agent guards, and escape procedures, detailing their functions and unblock methods.
Inputs & outputs
When to use wkharness-guards
- →Checking guard status
- →Unblocking development operations
- →Managing agent constraints
About this skill
Refresh:
wkappbot skill read wkharness-guards --if-newer— v1.171 (2026-05-15)
wkharness-guards
Steps
- GUARD CLUSTER G1-G6: G1 ASK-GUARD blocks ask run_in_background=true; G2 KILL-GUARD blocks terminate-cmdlet on wkappbot*; G3 SKILL-SEARCH-NUDGE soft nudge no block; G4 SKILL-EDIT-GUARD blocks .skill.json direct edit; G5 PACE-GUARD blocks Edit/Write when Sonnet%>Total%; G6 CP949-GUARD blocks non-CP949 chars in Write. -> unblock: wkappbot skill read wkharness-guards-howto
- GUARD CLUSTER G7-G12: G7 AGENT-BRIEF-GUARD blocks Agent with <3 skill reads; G8 OPUS-BUDGET-GUARD blocks Opus at 95%+ budget; G9 STALL-GUARD blocks 300s+10calls solo loop; G10 BG-TASK-GUARD blocks when bg mark >30s; G11 AGENTS-HEAL-GUARD auto-heals AGENTS.md once/session; G12 COMMIT-HOOK-GUARD auto-installs CP949 git hook. -> unblock: wkappbot skill read wkharness-guards-howto
- AGENT GUARDS: SPEC-GATE blocks Opus Agent without SPEC 1.Goal 2.Files 3.Approach 4.Constraints 5.Exit. CLAUDE-MD-SYNC blocks Agent without ## Pending in project MD. USAGE-TEST-GUARD 50pct Opus redirect to wkclaude.sh. WKCLAUDE-DIRECT blocks direct wkclaude.sh/cmd/ps1 calls.
- DEADLOCK ESCAPE: Delegation guards fire on Agent/Agent.cmd/wkcodex/wkclaude path ONLY. When blocked: DO WORK DIRECTLY with wkappbot/git/gh/./wk* leading token. skill* ALWAYS exempt from all guards. file-edit wrapper doubly exempt from pace+stall.
- GUARD CASCADE ORDER: (1) skill* always exempt. (2) file-edit wrapper doubly exempt. (3) wkcodex.sh safe from stall. (4) CronCreate entries need session-start registration. (5) ./wk* camelCase only -- dash after wk breaks gate.
- SKILL-UPDATE-PENDING: wkappbot file write/edit (non-skill) arms lock. Clear: (1) write first, (2) skill edit to clear, (3) delegate with no intervening write. Read/Grep safe.
- -> Per-guard unblock steps + full escape procedure: wkappbot skill read wkharness-guards-howto
- -> RIQUA/history, Codex harness, regex patterns, edge cases: wkappbot skill read wkharness-guards-ref
When not to use it
- →When not working with `wkharness.ps1` or related `wkappbot` workflows
- →When the user does not need quick reference for guards
- →When the user wants to bypass guards without understanding them
Limitations
- →Provides quick reference, not in-depth explanation
- →Guards fire on Agent/Agent.cmd/wkcodex/wkclaude path ONLY
- →Does not provide full details, refers to `wkharness-guards-howto` and `wkharness-guards-ref` skills
How it compares
This skill offers a concise, one-liner reference for `wkharness` guards and their unblock methods, unlike detailed documentation.
Compared to similar skills
wkharness-guards side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| wkharness-guards (this skill) | 0 | 1mo | No flags | Intermediate |
| linkerd-patterns | 6 | 4mo | Review | Advanced |
| linux-production-shell-scripts | 7 | 6mo | Review | Intermediate |
| aws-advisor | 5 | 5mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by kiexpert
View all by kiexpert →You might also like
linkerd-patterns
wshobson
Implement Linkerd service mesh patterns for lightweight, security-focused service mesh deployments. Use when setting up Linkerd, configuring traffic policies, or implementing zero-trust networking with minimal overhead.
linux-production-shell-scripts
davila7
This skill should be used when the user asks to "create bash scripts", "automate Linux tasks", "monitor system resources", "backup files", "manage users", or "write production shell scripts". It provides ready-to-use shell script templates for system administration.
aws-advisor
tech-leads-club
Expert AWS Cloud Advisor for architecture design, security review, and implementation guidance. Leverages AWS MCP tools for accurate, documentation-backed answers. Use when user asks about AWS architecture, security, service selection, migrations, troubleshooting, or learning AWS. Triggers on AWS, Lambda, S3, EC2, ECS, EKS, DynamoDB, RDS, CloudFormation, CDK, Terraform, Serverless, SAM, IAM, VPC, API Gateway, or any AWS service.
engineering-skills
alirezarezvani
23 production-ready engineering skills covering architecture, frontend, backend, fullstack, QA, DevOps, security, AI/ML, data engineering, computer vision, and specialized tools like Playwright Pro, Stripe integration, AWS, and MS365. 30+ Python automation tools (all stdlib-only). Works with Claude Code, Codex CLI, and OpenClaw.
managing-network-policies
jeremylongshore
Execute use when managing Kubernetes network policies and firewall rules. Trigger with phrases like "create network policy", "configure firewall rules", "restrict pod communication", or "setup ingress/egress rules". Generates Kubernetes NetworkPolicy manifests following least privilege and zero-trust principles.
gh-actions-validator
jeremylongshore
Validate use when validating GitHub Actions workflows for Google Cloud and Vertex AI deployments. Trigger with phrases like "validate github actions", "setup workload identity federation", "github actions security", "deploy agent with ci/cd", or "automate vertex ai deployment". Enforces Workload Identity Federation (WIF), validates OIDC permissions, ensures least privilege IAM, and implements security best practices.