Context & AI engineering
25,610 rows, by stacks then stars
1,865 of these skill files have been read, by 1,328 distinct authors, and what they tell an agent counted. What Context & AI engineering authors agree on, and what they forbid
Agent boundary change reviewer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-boundary-control-review/skills/agent-boundary-change-reviewer Skill
0★ repoUse when a model, prompt, Skill, tool, MCP server, source, egress rule, trace rule, retention rule, approval path, or runtime changes what an agent can see, reach, log, reuse, infer, or do.
Agent egress allowlist reviewer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-boundary-control-review/skills/agent-egress-allowlist-reviewer Skill
0★ repoUse when an agent workflow needs internet access, browser access, package registry access, vendor APIs, webhooks, search tools, MCP servers, or any outbound network destination.
Agent trace redaction gatekeeper
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-boundary-control-review/skills/agent-trace-redaction-gatekeeper Skill
0★ repoUse when an agent trace, prompt, tool call, screenshot, retrieval result, or model output may become a log, memory item, eval case, ticket, incident note, training example, or vendor-visible telemetry record.
Tool permission manifest writer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-launch-evidence-review/skills/tool-permission-manifest-writer Skill
0★ repoUse when an agent is about to receive a browser, file handle, MCP server, API, database, memory store, message-sending tool, or workflow action and the allowed authority must be written before access.
Agent capability diff reviewer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-tool-authority-review/skills/agent-capability-diff-reviewer Skill
0★ repoUse when an updated agent, prompt, Skill, connector, tool, memory path, or retrieval path changes what the workflow can do.
Agent context quarantine mapper
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-tool-authority-review/skills/agent-context-quarantine-mapper Skill
0★ repoUse when an agent reads untrusted webpages, emails, documents, retrieval results, logs, issues, comments, or tool output before acting.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-tool-authority-review/skills/agent-permission-card-writer Skill
0★ repoUse when an agent, connector, MCP server, browser profile, repository workflow, or automation needs scoped permissions before tool access.
Agent service identity assigner
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-tool-authority-review/skills/agent-service-identity-assigner Skill
0★ repoUse when an agent acts through a human browser session, broad user token, shared credential, or service account.
Browser agent profile isolator
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-tool-authority-review/skills/browser-agent-profile-isolator Skill
0★ repoUse when a browser agent needs a profile, cookies, session state, extension access, downloads, or remote debugging access.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-tool-authority-review/skills/github-agent-input-reviewer Skill
0★ repoUse when an agent reads or acts on GitHub issues, pull requests, comments, workflow files, repo rules, code review text, or generated patches.
Permission lease record writer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-tool-permission-lease-review/skills/permission-lease-record-writer Skill
0★ repoUse when an agent, subagent, MCP server, browser profile, repository workflow, connector, or automation needs a temporary tool permission record before access is enabled.
Agent acceptance criteria setter
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-workflow-control-review/skills/agent-acceptance-criteria-setter Skill
0★ repoUse when an agent task needs done criteria, evidence rules, source rules, tool rules, blocked outputs, and verification before work starts.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-workflow-control-review/skills/agent-action-preview-builder Skill
0★ repoUse when an agent proposes a write, send, deploy, publish, payment, CRM update, permission change, or other side effect.
Agent approval packet assembler
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-workflow-control-review/skills/agent-approval-packet-assembler Skill
0★ repoUse when a human reviewer needs the evidence packet for approving, rejecting, or escalating an agent action.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-workflow-control-review/skills/agent-retry-budget-setter Skill
0★ repoUse when an agent workflow needs retry limits, replanning limits, stop conditions, escalation triggers, and irreversible-action locks.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/agent-workflow-control-review/skills/agent-source-labeler Skill
0★ repoUse when agent claims, proposed actions, summaries, or decisions need source identity before the workflow relies on them.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/ai-workflow-eval-and-lifecycle-review/skills/negative-skill-eval-writer Skill
0★ repoUse when a reusable Skill, prompt, agent, or workflow needs misuse, missing-context, hostile-input, expanded-tool, and changed-model eval cases before publication.
Prompt injection workflow tester
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/ai-workflow-eval-and-lifecycle-review/skills/prompt-injection-workflow-tester Skill
0★ repoUse when a workflow reads untrusted webpages, issues, emails, documents, retrieval content, attachments, or tool output before acting.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/handoff-ownership-transfer-review/skills/blocked-handoff-repair-router Skill
0★ repoUse when a handoff packet is incomplete, unsafe, unsupported, stale, prompt-injected, or rejected by the receiving owner.
Ownership handoff packet writer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/handoff-ownership-transfer-review/skills/ownership-handoff-packet-writer Skill
0★ repoUse when a current owner needs to prepare a compact packet before another agent, human, orchestrator, or service accepts responsibility for the work.
Memory expiry and rollback planner
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/memory-write-quarantine-review/skills/memory-expiry-and-rollback-planner Skill
0★ repoUse when a memory item needs retention rules, expiry, review cadence, snapshot handling, deletion path, rollback owner, and audit evidence before durable storage.
Memory poisoning test designer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/memory-write-quarantine-review/skills/memory-poisoning-test-designer Skill
0★ repoUse when a workflow needs a cross-session test that checks whether hostile, stale, low-trust, or unsupported memory can steer future planning, retrieval, tools, or output.
Memory proposal intake reviewer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/memory-write-quarantine-review/skills/memory-proposal-intake-reviewer Skill
0★ repoUse when a raw or planned memory write needs source labels, evidence handling, memory type, and quarantine status before it can become durable.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/memory-write-quarantine-review/skills/memory-sensitivity-gatekeeper Skill
0★ repoUse when proposed memory may include personal data, customer data, credentials, private URLs, regulated data, permission claims, approval claims, security posture, or internal-only context.
Memory trust and influence scorer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/memory-write-quarantine-review/skills/memory-trust-and-influence-scorer Skill
0★ repoUse when proposed memory needs a trust score, confidence note, contradiction review, and allowed influence boundary before future agents can retrieve it.
Prompt injection denial scenario writer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/negative-authorization-test-review/skills/prompt-injection-denial-scenario-writer Skill
0★ repoUse when untrusted source text can reach an agent with tools and the team needs denial scenarios that combine a benign task with hostile email, web, issue, document, memory, MCP, or tool-result content.
Prompt change decision recorder
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/prompt-mismatch-log-review/skills/prompt-change-decision-recorder Skill
0★ repoUse when reviewers need to decide and document whether to rewrite a prompt, edit a Skill, add an example, change schema, change retrieval, change evaluator, add approval, or leave unchanged.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/sales-to-implementation-handoff/skills/crm-safe-handoff-summary Skill
0★ repoUse when reviewed sales-to-implementation handoff context needs to be saved into CRM or project systems while excluding sensitive details, unsupported commitments, and internal sales strategy.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/sales-to-implementation-handoff/skills/handoff-input-safety-check Skill
0★ repoUse when closed-won context, sales notes, commitments, customer data, or implementation assumptions need screening before handoff planning.
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/workflow-failure-label-review/skills/eval-case-converter Skill
0★ repoUse when a recurring failure label should become an eval scenario with expected safe behavior, blocked behavior, evidence boundary, approval route, and critical failure conditions.
Agent resume checkpoint reviewer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/workflow-readiness-review/skills/agent-resume-checkpoint-reviewer Skill
0★ repoUse when a paused, scheduled, interrupted, failed, or checkpointed AI agent may resume from saved state, pending tool calls, approvals, traces, or memory.
Prompt review checkpoint writer
vibesec-advisory/vibesec-advisory-skill-library/SKILLS/workflow-readiness-review/skills/prompt-review-checkpoint-writer Skill
0★ repoUse when an AI-generated plan, draft, answer, or recommendation needs a human checkpoint that preserves judgment before the output becomes workflow state.
commentary-dev/commentary-skills/plugins/commentary-review/skills/commentary-draft-review Skill
0★ repoUse Commentary draft reviews from an agent via the Commentary CLI for live collaborative review of Markdown, MDX, HTML, and plain text artifacts. Trigger this skill when a user asks to create, draft, edit, revise, share, or review a blog post, spec, proposal, report, documentation page, or other text artifact with Commentary; when they mention Commentary draft reviews, live comments, watch-comment, review it with commentary, or using an agent and human reviewer together; and when the agent needs to set up authentication, create/sync/watch/share a review, read or wait for comments, reply to comments, resolve threads, manage comments by other reviewers, or decide whether to use the CLI or an already-installed Commentary MCP server.
Commentary live preview review
commentary-dev/commentary-skills/plugins/commentary-review/skills/commentary-live-preview-review Skill
0★ repoSet up and use Commentary Live Preview Reviews for web apps through the Commentary review SDK and MCP. Trigger this skill when a user wants to add @commentary-dev/review-sdk or the Commentary CDN SDK to a React, Next.js, Vite, Vue, Svelte, Angular, Astro, static HTML, or other browser app; launch a localhost or deployed live preview review; listen for Web App Review comments through the Commentary MCP server; address selected-element comments; resolve review threads; handle localhost immediate feedback; or apply deployed-review rules based on review owner, current user, comment author, and agent alias tags.
commentary-dev/commentary-skills/skills/commentary-draft-review Skill
0★ repoUse Commentary draft reviews from an agent via the Commentary CLI for live collaborative review of Markdown, MDX, HTML, and plain text artifacts. Trigger this skill when a user asks to create, draft, edit, revise, share, or review a blog post, spec, proposal, report, documentation page, or other text artifact with Commentary; when they mention Commentary draft reviews, live comments, watch-comment, review it with commentary, or using an agent and human reviewer together; and when the agent needs to set up authentication, create/sync/watch/share a review, read or wait for comments, reply to comments, resolve threads, manage comments by other reviewers, or decide whether to use the CLI or an already-installed Commentary MCP server.
Commentary live preview review
commentary-dev/commentary-skills/skills/commentary-live-preview-review Skill
0★ repoSet up and use Commentary Live Preview Reviews for web apps through the Commentary review SDK and MCP. Trigger this skill when a user wants to add @commentary-dev/review-sdk or the Commentary CDN SDK to a React, Next.js, Vite, Vue, Svelte, Angular, Astro, static HTML, or other browser app; launch a localhost or deployed live preview review; listen for Web App Review comments through the Commentary MCP server; address selected-element comments; resolve review threads; handle localhost immediate feedback; or apply deployed-review rules based on review owner, current user, comment author, and agent alias tags.
Traigent/traigent-skills/skills/traigent-boost-agent Skill
0★ repoEnd-to-end lifecycle playbook — from a single decorated function to a full 12-step codebase onboarding — for adding Traigent to an existing client agent codebase and measurably boosting accuracy, cost, latency, or reliability. Use when asked to add Traigent to an existing agent codebase, onboard this agent to Traigent end-to-end, run a full agent-build lifecycle, wire an evaluator and optimize for an agent with code to instrument, boost accuracy/cost of an existing agent, select TVARs with generate_config() on agent code, choose composite knobs by agent shape, instrument @traigent.optimize into agent code, validate in mock mode, run real optimization with budgets, inspect results, iterate to refine an existing agent, or gate a promoted config. For brand-new users without an existing agent codebase, see traigent-setup-quickstart. ALWAYS start with dry-run (mock mode) to validate the full pipeline, then switch to real execution only when the user explicitly requests it.
Traigent/traigent-skills/skills/traigent-eval-audit Skill
0★ repoAgent skills for deploy and use Traigent
Traigent/traigent-skills/skills/traigent-eval-build Skill
0★ repoAgent skills for deploy and use Traigent
Traigent/traigent-skills/skills/traigent-eval-choose-metric Skill
0★ repoChoose Traigent objectives and metric functions before optimizing. Use when asked which metric to use, how to measure quality, whether to optimize accuracy/cost/latency/safety, how to name objectives, how to combine multiple objectives, when to use custom metric_functions, or how to turn product goals into Traigent objectives.
gaguero/seo-agent-skills/seo-orchestrator Skill
0★ repoPlans and routes SEO work for AI agents. Takes a URL or site plus a business goal, language, market and competitors, then produces a prioritized, impact/effort/risk-ranked SEO plan and dispatches the right specialist skills (technical audit, content brief, schema, AI-search, quality gates). Use this first whenever someone asks to "do SEO", "audit my site", "improve rankings", "optimize a page", "rank in AI / ChatGPT / AI Overviews", or when the request spans more than one SEO discipline. Coordinates the rest of the seo-agent-skills pack.
mysteryon88/skills/mina-protocol/skills/mina-protocol-agent Skill
0★ repoUse when installing, choosing, coordinating, or operating the Mina Protocol/o1js skill pack, especially when shared playbooks, examples, templates, source references, or multiple Mina specialist skills may be needed.
JimmyBlanquet/project-forge/skills/ralph-agent-teams Skill
0★ repoParallel implementation of Ralph++ PRD sub-stories using Claude Opus 4.6 Agent Teams. Use when executing prd.json with multiple independent sub-stories that can run in parallel. Extends ralph-loop with multi-agent orchestration, dependency analysis, conflict resolution, and progress synchronization. Requires CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1.
webdevtodayjason/dev-kit/plugins/dev-kit/skills/llm-council Skill
no license0★ repoMulti-LLM collaborative brainstorming and planning. Use when user explicitly requests consultation with multiple AI models (ChatGPT, Gemini, other LLMs) before presenting an implementation plan, or asks to "consult the council", "ask other models", or "get perspectives from other AIs". Queries external LLM APIs, synthesizes their perspectives, and presents an adapted implementation plan.
The same file is in 8 other repositories
webdevtodayjason/dev-kit/plugins/dev-kit/skills/setup-matt-pocock-skills Skill
no license0★ repoSets up an `## Agent skills` block in AGENTS.md/CLAUDE.md and `docs/agents/` so the engineering skills know this repo's issue tracker (GitHub or local markdown), triage label vocabulary, and domain doc layout. Run before first use of `to-issues`, `to-prd`, `triage`, `diagnose`, `tdd`, `improve-codebase-architecture`, or `zoom-out` — or if those skills appear to be missing context about the issue tracker, triage labels, or domain docs.
The same file is in 185 other repositories
webdevtodayjason/dev-kit/plugins/dev-kit/skills/write-a-skill Skill
no license0★ repoCreate new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.
The same file is in 20 other repositories
giostriquer/agent-workshop/plugins/agent-workshop/skills/agent-workshop-onboard Skill
0★ repoUse when onboarding Agent Workshop agents into a repository, planning or applying repo-local agent packs, auditing existing agent-workshop adoption, or explaining pack/profile choices.
giostriquer/agent-workshop/plugins/toolkit/skills/handoff-goal Skill
0★ repoUse when a goal — the rest of an in-flight plan, a scoped slice of it, or a brand-new idea — should be handed to a new session to pursue autonomously, or when an existing goal contract needs critique. Produces a self-contained goal contract directory (`tmp/<YYYY-MM-DD>-<goal-slug>/`) holding `goal.md`, the frozen contract (outcome, baseline, acceptance checks with a real-surface primary verifier, integrity rules, approval gates, operating rules), and `plan.md`, the living route (phases, bounded ledger, next action) the pursuing session maintains, with history archived to `ledger.md` — so pursuit stays consistent across compactions on any runtime, including Codex goal mode, which can activate the contract directly. With no argument it infers the goal from session context and confirms; pointed at an existing goal directory (or "critique") it audits and tightens instead of writing anew; never pursues the goal itself.
giostriquer/agent-workshop/plugins/toolkit/skills/handoff-pr Skill
0★ repoUse when a branch is ready for a PR but the current session is not authorized to open one. Produces a structured PR handoff artifact (title, body, ticket links, validation and review status) for a separately-authorized session or agent to open. Derives the PR body from the repo's own PR template when one exists. Auto-detects the ClickUp / Linear / Jira ticket and asks to confirm. Writes the artifact to a `tmp/` scratch file by default; `inline` prints it in-session instead and writes no file. Never opens the PR itself.
giostriquer/agent-workshop/plugins/toolkit/skills/handoff-review Skill
0★ repoUse when work on a branch needs a fresh, unbiased pair of eyes — a pre-PR review, or a clean restart when this session's context has gone bad and a new session should independently verify what was already done and continue it. Produces a self-contained brief (task-vs-code, rules conformance, information leak, correctness) that re-derives the task from ticket + diff, never trusting the prior session. Default spawns a fresh reviewer (verify-only); `handoff`/`session` writes the brief to a scratch file for a new session (verify-only); `continue`/`resume` writes a verify-then-continue brief carrying current state, remaining work, and operating rules, pointing to handoff-goal for substantial forward work.
fxjim/agent-commerce-guard/plugins/agent-commerce-guard/skills/agent-commerce-action-review Skill
no license0★ repoReview proposed AI-agent actions locally before execution. Use for deployments, external publishing, financial actions, marketplace jobs, social actions, destructive commands, billing changes, elevated permissions, or credential-dependent work. Produce an allow, review, or deny decision with reasons and required evidence. Never request or reproduce secrets, customer data, credentials, or private wallet material.
fxjim/agent-commerce-guard/skills/buy-agent-commerce-guard Skill
no license0★ repoInspect and buy the Agent Commerce Guard 1 USDC Base launch pass through an authorized Base MCP wallet, x402 client, or browser wallet. Use when a user wants to evaluate the offer, confirm payment details, purchase the package, verify a Base transaction, or unlock the paid download. Require explicit approval before any payment and never request wallet secrets.
fxjim/agent-commerce-guard/skills/try-agent-commerce-guard Skill
no license0★ repoEvaluate non-sensitive AI-agent commerce action manifests with the free Agent Commerce Guard hosted policy endpoint. Use when a user wants to preflight wallet spend, paid API calls, deployments, token launches, marketplace jobs, social actions, destructive commands, billing changes, or credential access before execution. Never send secrets, private source, customer data, or credentials to the hosted demo.
aidehua08/codex-obsdian/plugins/codex-obsdian/skills/wiki-retrieve Skill
0★ repoBuild and query a vault-local contextual BM25 retrieval index with optional multilingual Nomic cosine reranking; use for retrieve, hybrid retrieval, BM25, rerank, contextual retrieval, chunk search, vault search, semantic search, find relevant passages, or retrieval diagnostics. Derived caches stay under .vault-meta, remote egress requires explicit consent, and unavailable reranking falls back deterministically.
The same file is in 1 other repository
aidehua08/codex-obsdian/skills/wiki-retrieve Skill
0★ repoBuild and query a vault-local contextual BM25 retrieval index with optional multilingual Nomic cosine reranking; use for retrieve, hybrid retrieval, BM25, rerank, contextual retrieval, chunk search, vault search, semantic search, find relevant passages, or retrieval diagnostics. Derived caches stay under .vault-meta, remote egress requires explicit consent, and unavailable reranking falls back deterministically.
The same file is in 1 other repository
evertonfernandes3321-wq/mythos-skills/skill-authoring Skill
0★ repoMeta-skill para criar e revisar skills/superprompts de alta qualidade para agentes de IA, em qualquer dominio e stack — seletor de padrao (Tool Wrapper / Pipeline / Generator / Reviewer / Inversion), frontmatter correto (name=pasta, description quando+o-que <1024 char 1 linha), progressive disclosure com references, estilo Mythos (persona de elite, escopo stack-agnostico, checklist sub-atomico, formato de saida fixo, auto-verificacao) e teste com contexto fresco. Use para autorar uma skill nova, padronizar uma familia de skills, ou revisar/auditar a qualidade de uma skill existente antes de publicar.
glim-sh/skills Skill
0★Live data for AI agents - Twitter/X search, Reddit posts and comments, semantic web search, clean page extraction, GitHub repos/code/PRs, Amazon products, YouTube transcripts. One hosted MCP server at glim.sh/mcp. Sign in with OAuth and a prepaid balance, or pay per call with crypto (x402 USDC on Base/Solana, MPP on Tempo). No API keys to provision.
lensetek/Startup-Agent-Skills-Hub/skills/startup-workflow-orchestrator Skill
0★ repoThe Startup Workflow Orchestrator selects the right specialist agents, sequences handoffs, enforces security gates, and keeps startup delivery workflows aligned from strategy through launch.
tjrice42/design-judgment-skills/handoff-spec Skill
no license0★ repoGenerate a complete dev-handoff spec with an exhaustive edge-case and state matrix for any prototype, design, or feature. Use this skill whenever the user finishes or shares a prototype, mentions handing off to dev/engineering, asks about edge cases, states, error states, empty states, loading states, "what am I missing", design QA, or wants a spec written. Also trigger proactively at the end of any prototype-building session — a prototype is not done until its spec exists. Even if the user only says "wrap this up" or "get this ready for the team," use this skill.
tjrice42/design-judgment-skills/prompt-retro Skill
no license0★ repoAnalyze how the user has been prompting Claude — from build journals, session transcripts, or pasted prompts — and run a retrospective showing which prompts worked, which cost extra iterations, why, and how to prompt better next time, with rewritten examples of the user's actual prompts. Use this skill whenever the user asks how their prompting is going, wants prompting feedback or insights, says "how could I have prompted that better," wants to improve their AI workflow, or at the natural end of a build session as a closing retro. Maintains a cumulative personal prompting playbook across sessions.