skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
oakoss/agent-skills95 installs

agent-standards

Defines behavioral and cognitive standards for senior AI engineering agents. Use when configuring agent reasoning protocols, memory management, or context engineering strategies. Use for autonomous reasoning, tiered memory systems, verifiable goal execution, multi-agent orchestration, and token optimization.

How do I install this agent skill?

npx skills add https://github.com/oakoss/agent-skills --skill agent-standards
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    The agent-standards skill provides cognitive and architectural guidelines for AI agents, including reasoning stacks, memory management, and context engineering. It contains no executable code and explicitly directs agents to follow security best practices such as secret scrubbing.

  • Socketpass

    No alerts

  • Snykpass

    Risk: LOW · No issues

  • Runlayerwarn

    5/5 files flagged

What does this agent skill do?

Expert Instruction

Overview

Foundational meta-skill that defines behavioral and cognitive standards for senior AI engineering agents. Establishes the reasoning pipeline, memory architecture, and context engineering practices that enable autonomous, long-horizon task execution with verifiable outcomes.

When to use: Configuring agent reasoning, managing context windows, establishing verification protocols, orchestrating multi-agent workflows, optimizing token usage.

When NOT to use: Domain-specific coding tasks (use specialized skills), UI/UX design, database schema work.

Quick Reference

PatternApproachKey Points
PerceptionAnalyze terminal output, codebase, tracesHigh-fidelity input ingestion
HypothesisGenerate multiple solution pathsEvaluate before committing
SimulationReason through change consequencesPredict side effects
ActionPrecise tool executionAtomic, testable commits
CriticismSelf-audit outputCheck for bugs and style violations
Context discoveryMap framework versions and patternsAlways discover before implementing
Dependency auditCheck existing tools before adding new onesAvoid unnecessary dependencies
Verifiable planningDefine Definition of DoneTest pass, build success, or user approval
Interactive alignmentAsk the user for ambiguous requirementsConfirm critical architectural decisions
Atomic implementationApply changes in logical, testable unitsEach commit should be independently verifiable
Audit and cleanupRun linter, remove debug artifactsNo temporary code in final output
Selective readingUse offset and limit parametersAvoid reading entire large files
Symbol searchUse grep/rg to find definitionsDo not read entire directories
Few-shot anchoringProvide canonical examplesMore effective than long rule lists
Memory tieringShort-term, mid-term, long-termMatch persistence to information lifetime
Context packingBundle related filesStructured markdown artifacts
Noise reductionExclude node_modules, dist, binary artifactsMaximize signal-to-noise ratio in context
Semantic summarizationCondense long logs into actionable factsSingle-sentence failure descriptions
Cognitive load pruningRemove irrelevant history from active contextFree tokens for current task reasoning

Common Mistakes

MistakeCorrect Pattern
Failing silently when a tool call or build step errorsAlways report status and errors explicitly to the user
Inventing APIs or methods that do not existSearch documentation or use web search to verify API signatures before using them
Writing verbose explanations instead of showing codePrioritize code-first communication; explain only when asked
Ignoring surrounding code style and conventionsMimic the existing codebase patterns, naming, and formatting
Hardcoding secrets or API keys in source filesUse environment variables and .env file mapping
Reading entire directories to find a single symbolUse grep or rg to locate definitions, then read only relevant sections
Skipping verification after implementationEvery task must have a verification signal before marking complete
Storing sensitive data in memory or context filesRun a secret scrub before persisting any memory vector
Loading full file contents into context unnecessarilyUse partial reads with offset and limit for large files
Including duplicate information from multiple sourcesDeduplicate context to preserve token budget

Delegation

  • Explore a codebase to map framework versions and active patterns: Use Explore agent
  • Execute a complex multi-step implementation with atomic commits: Use Task agent
  • Plan architecture for a long-horizon feature with dependency analysis: Use Plan agent

References

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/oakoss/agent-skills/agent-standards">View agent-standards on skillZs</a>