SkillPedia Skills Encyclopedia
中文 EN
Today's updates 2026-11-18 UTC

No new commits yet today

Open the update feed

agents-best-practicesDenisSergeevitch/agents-best-practices

Provides a practical engineering checklist for designing AI agent workflows, turning vague questions like how autonomous…

No ratings yet Rate this skill
Stars2.4k
Forks216
Watchers2.4k
Open Issues7

InstallOne-line install

Installable · 19 skills

This repository follows the Agent Skills spec, so one command installs it into your AI client — works across Claude Code, Codex, Cursor, Copilot, Windsurf, and 18+ other clients.

npx skills@latest add DenisSergeevitch/agents-best-practices

19 more are not listed — run npx skills@latest add DenisSergeevitch/agents-best-practices --list to see them all.

Add -g install globally (available in every project) · add --agent claude-code to target a client · add --yes to skip the prompts. Restart your AI client afterward and it will pick it up.

OverviewOverview

Key Features

Provides a practical engineering checklist for designing AI agent workflows, turning vague questions like how autonomous an agent should be, which tools it may touch, and how failures are contained into item-by-item design decisions. Key capabilities include: 1. Domain and autonomy leveling: define the work domain the agent handles, then set its autonomy level to decide how far it may act on its own, avoiding over- or under-privileging; 2. Risk-driven loop: adjust the agent's execution loop by risk level, routing high-risk actions through stricter confirmation and rollback paths while low-risk actions stay fast; 3. Tool surface and call-result constraints: define the tool surface available to the agent and require every tool call to return an explicit result, eliminating silent failures and dangling calls; 4. Draft and commit separation: split changes into a Draft stage for review and a Commit stage for execution, reducing the risk of unintended modifications; 5. One-time upstream check and validation: verify and pin upstream dependencies once, and use a Validation step to uniformly check agent output so results stay reproducible.

TopicsTopics

RatingUser ratings

Nobody has rated this skill yet. Yours would be the first on the site — your vote becomes the initial average.

Rate this skill

Fill it in to be included in the per-role averages below.

One vote per IP,and it cannot be changed after submitting, so choose carefully the first time. No account is collected and no sign-in is required.

RepositoryRepository

Skill ID406
Data sourceGitHub
RepositoryDenisSergeevitch/agents-best-practices
CategoryAI Agent Development / Agent Engineering & Workflow Design
Primary languageOther
LicenseMIT License
Default branchmain
GitHub repository ID1240191065
Node IDR_kgDOSevQWQ
Rank#403 (position within the source ranking)
Site-wide popularity rank#473 (computed live from Stars, across all 5833 entries)

GrowthStar / Fork growth

One snapshot per day, tracking how Stars and Forks changed

01k2k3k2026-09-292026-10-112026-09-29 · 2,3612026-10-11 · 2,378
  • Now2,378
  • Previous point · 09-29 2,361
  • vs previous point +17
  • 2 data points 12 days
01002003002026-09-292026-10-112026-09-29 · 2132026-10-11 · 216
  • Now216
  • Previous point · 09-29 213
  • vs previous point +3
  • 2 data points 12 days

TimelineTimeline

Repository created2026-05-15 21:32:216 months ago
Data last updated2026-10-26 11:57:3523 days ago
Last commit2026-10-05 00:16:421 months ago

CommitLatest commit

Committed by DenisSergeevitch. Update Category: Documentation Updates. Commit details: Document always-on agents and durable runtime contracts. This update refines the documentation for easier onboarding.

28e87b5 DenisSergeevitch 2026-10-05

Next stepsNext steps