claude

devagent-standards

Engineering standards as fourteen on-demand skills: auth, API, browser and LLM security; encryption; cloud infrastructure and cost; stack selection; observability; document artifacts; breach response; verification methods; engagement close-out; public release; feature flags. Plus devagent-guard hooks enforcing the toolchain rules (deny or ask).

1 on GitHubby DevAgent contributorsv1.1.0UNLICENSEDupdated 23 days agosecurity
/

Plugin devagent-standards from marketplace devagent in mrjones-cs/devagent. Commands read from the repo's manifest — 23 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (14)

  • api-security

    Standard for APIs and any machine-facing endpoint. Use when building, exposing, reviewing, or deploying an API - REST, GraphQL, webhooks, Lambda/Function URLs, API Gateway, ALB or container ingress, service-to-service calls, API keys, rate limits, pagination, or multi-tenant data access. Also use wh

  • auth-portal

    Standard for any system with a login. Use when building, reviewing, deploying, or debugging authentication - login pages, signup, session handling, MFA enrolment or challenges, password reset and account recovery, JWT or cookie sessions, OAuth/OIDC/SAML, Cognito, Entra ID, Auth0, Identity Center, or

  • cloud-infra

    Standard for public cloud work. Use when provisioning, modifying, deploying, or reviewing cloud infrastructure on AWS/GCP/Azure/Cloudflare - Terraform/OpenTofu/Pulumi/CloudFormation/CDK/Bicep, accounts and OUs, IAM, VPCs, budgets and cost estimates, teardown scripts, or deploying a demo or client-fa

  • document-artifact

    Standard for generated document artifacts. Use whenever creating or regenerating a PDF, DOCX, PPTX, slide deck, one-pager, capability statement, brochure, white paper, case study, proposal, or an image rendered from one - by any method (python-pptx, reportlab, fpdf2, python-docx, Markdown-to-PDF, he

  • encryption

    Standard for encryption in transit and at rest. Use when creating or modifying any data store (bucket, database, volume, cache, queue, search index, warehouse, backup or snapshot), configuring TLS or certificates, choosing KMS/CMK vs provider-managed keys, hashing passwords, handling secrets on disk

  • engagement-closeout

    Standard for ending an engagement. Use when a project is finishing, delivering, or being handed over - final delivery and acceptance, transferring repos, IaC state, secrets and cloud accounts, revoking access in both directions, license and IP attestation or an SBOM at delivery, returning or destroy

  • feature-flags

    Standard for feature flags, toggles, kill switches and gradual rollouts in software we build. Use when adding, evaluating, flipping or removing a flag - server-side versus client-side evaluation, client SDK keys and bootstrap payloads, flagged-off code in public bundles, who may change a flag and ho

  • incident-breach

    Playbook for a suspected breach of a customer-facing system - unauthorized access, data exposure, compromised account or key with blast radius, ransomware, or any event that may trigger a contractual notification clock. Use when a breach is suspected or confirmed, when assessing notification obligat

  • llm-app-security

    Standard for AI-powered software we build. Use when designing or reviewing anything that calls a model - prompts, system prompts, agents, tool/function calling, RAG and retrieval, embeddings and vector stores, chat UIs, model output that reaches a shell/SQL/renderer, model vendor selection, or promp

  • observability

    Standard for making a system operable. Use when designing or handing over a customer-facing system - structured logging, correlation IDs, health/liveness/readiness endpoints, metrics, tracing, alerting and thresholds, availability/latency/durability targets, backups and restore testing (RTO/RPO), da

  • public-release

    Standard for anything about to become public. Use before publishing, launching or deploying a website, landing page, blog post, case study, portfolio or marketing page; before making a repository public, open-sourcing code, posting a gist, or sharing a screenshot, recording or demo outside the firm;

  • stack-selection

    Standard for choosing a language and runtime version. Use at the start of a greenfield project, when proposing a stack, when adding a second language or runtime to a delivery, when picking a base image or managed service version, when setting up the project check command and type checking, or when a

  • verification-methods

    Method for proving a built thing actually works before reporting it done. Use when finishing, validating, testing, or about to claim completion on ANY work product - code, a PDF/DOCX/PPTX or slide deck, a diagram or chart, a web page or UI change, a deployed endpoint, or cloud infrastructure. Covers

  • web-browser-security

    Standard for anything that renders in a browser. Use when building or reviewing a web page, SPA, or front end - Content Security Policy, security headers, XSS, CSRF, cookies, clickjacking, third-party scripts and analytics tags, file uploads, or rendering user-supplied or model-generated HTML/Markdo

Hooks (3)

  • PreToolUse

    python3 "${CLAUDE_PLUGIN_ROOT}/hooks/guard.py" pre

  • PostToolUse

    python3 "${CLAUDE_PLUGIN_ROOT}/hooks/guard.py" post

  • SessionStart

    python3 "${CLAUDE_PLUGIN_ROOT}/hooks/guard.py" session

View source on GitHub

More from mrjones-cs/devagent

All 2 plugins →

delivery

claude

A delivery workflow: load, brainstorm, design, research, spec-panel, workflow, build, plan and save, each writing a markdown record of one feature. Build executes the plan one task at a time and ticks a task only with recorded verification evidence.

8 skills · 4 agents · hooks

1

Similar plugins

ecc

claudecodex

Harness-native ECC operator layer - 71 agents, 314 skills, 98 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses

200 skills · 98 commands · 71 agents · 1 MCP server · hooks

affaan-m/ECC 302,088

Security review for Claude-generated code. Pattern-based warnings on edits, LLM-powered diff review on Stop, and an agentic commit reviewer that catches injection, XSS, SSRF, hardcoded secrets, and 25+ other vulnerability classes.

hooks

anthropics/claude-code 155,476

CMEM Pro

claude

CMEM Pro for the Claude app - hooks stream tool use to cmem.ai and inject observations into new sessions and agents

3 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

Agent runtimes — local WASM-sandboxed agents (rvagent) + Anthropic Claude Managed Agents (cloud); one interface, local-vs-cloud backends

5 skills · 2 commands · 9 agents

ruvnet/ruflo 81,589

Install the "AAS DevOps & Cloud" workflow plugin from Agentic Awesome Skills.

18 skills

sickn33/agentic-awesome-skills 49,620

Cloud architecture design for AWS/Azure/GCP/OCI, Kubernetes cluster configuration, Terraform infrastructure-as-code, hybrid cloud networking, and multi-cloud cost optimization

8 skills · 7 agents

wshobson/agents 42,461

Maintain devagent-standards?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins2 plugins