Context & AI engineering
25,815 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
Per agent skill listing dedupe
ychampion/cskill-agents/skills/per-agent-skill-listing-dedupe Skill
36★ repoKeep skill-listing deltas scoped per agent or thread so each agent gets its own first-turn announcement surface.
The same file is in 1 other repository
Plugin only agent surface gating
ychampion/cskill-agents/skills/plugin-only-agent-surface-gating Skill
36★ repoSkip agent-declared MCP servers when strict plugin-only mode is active unless the agent is admin-trusted, keeping the session secure.
The same file is in 1 other repository
Prevent continuation stop hook contract
ychampion/cskill-agents/skills/prevent-continuation-stop-hook-contract Skill
36★ repoReturn a narrow stop-hook result that can halt the query loop without throwing, while still surfacing blocking errors separately.
The same file is in 1 other repository
Prevent continuation stop hook messaging
ychampion/cskill-agents/skills/prevent-continuation-stop-hook-messaging Skill
36★ repoCollect stop-hook progress/errors, emit a structured prevented-continuation attachment, and summarize the block for users and logs.
The same file is in 1 other repository
ychampion/cskill-agents/skills/progressive-compaction-prompt Skill
36★ repoDrive compaction summaries through structured prompts that sequence naming, artifacts, and invocation guidance before stripping analysis tags.
The same file is in 1 other repository
Prompt context handoff for agent runtime
ychampion/cskill-agents/skills/prompt-context-handoff-for-agent-runtime Skill
36★ repoPackage prompt, identity, and runtime constraints into one explicit handoff so child agents start with the right context and no accidental leakage.
The same file is in 1 other repository
Relevant memory selection by header
ychampion/cskill-agents/skills/relevant-memory-selection-by-header Skill
36★ repoSelect the most useful memory files by comparing headers/descriptions via a JSON-schema-guided assistant check.
The same file is in 1 other repository
Reserved summary output budget
ychampion/cskill-agents/skills/reserved-summary-output-budget Skill
36★ repoReserve part of the context window for summary output so proactive compaction does not kill the response budget.
The same file is in 1 other repository
ychampion/cskill-agents/skills/role-aware-context-slimming Skill
36★ repoTrim user and system context based on the agent\u2019s role so read-only helpers don\u2019t inherit stale git or CLAUDE metadata.
The same file is in 1 other repository
ychampion/cskill-agents/skills/safe-worktree-slug-validation Skill
36★ repoEnsure Git worktree slugs can never escape the managed workspace or inject traversal.
The same file is in 1 other repository
Session memory first compaction
ychampion/cskill-agents/skills/session-memory-first-compaction Skill
36★ repoAttempt session-memory compaction before full conversation compaction so lightweight summaries win when available.
The same file is in 1 other repository
Session memory mailbox exclusion
ychampion/cskill-agents/skills/session-memory-mailbox-exclusion Skill
36★ repoPrevent session-memory background forks from draining teammate mailbox attachments that belong to the main lead thread.
The same file is in 1 other repository
Slash command skill classification filter
ychampion/cskill-agents/skills/slash-command-skill-classification-filter Skill
36★ repoDerive a slash-command skill surface from a broader prompt-command registry by requiring descriptive metadata, excluding builtins, and admitting only a controlled set of origins plus explicit user-only commands.
The same file is in 1 other repository
ychampion/cskill-agents/skills/subagent-metrics-bridging Skill
36★ repoTranslate a streaming subagent run into task progress, analytics, and notifications while keeping partial results and cleanup tidy.
The same file is in 1 other repository
Token budgeted compaction with reinjection
ychampion/cskill-agents/skills/token-budgeted-compaction-with-reinjection Skill
36★ repoKeep compaction budgets under control by compacting incrementally and reinjecting truncated attachments only when the reasoning surface remains stable.
The same file is in 1 other repository
Tool success aware memory suppression
ychampion/cskill-agents/skills/tool-success-aware-memory-suppression Skill
36★ repoSuppress context docs for tools that already succeeded in the current turn while keeping help available for failures or unresolved runs.
The same file is in 1 other repository
Turn scoped memory prefetch with disposal
ychampion/cskill-agents/skills/turn-scoped-memory-prefetch-with-disposal Skill
36★ repoStart a non-blocking, turn-scoped memory prefetch that auto-disposes and only injects results once per turn.
The same file is in 1 other repository
MoizIbnYousaf/marketing-cli/skills/exa-contents Skill
35★ repoCall Exa Contents (POST /contents) for LLM-ready extraction from known URLs: text, highlights, summaries, links, image links, subpages, freshness-controlled crawl. Use when the agent already has URLs. Prefer Exa MCP web_fetch_exa when available; otherwise raw HTTP with EXA_API_KEY. NOT for open-ended discovery (use exa-search) or auth-walled X/Twitter (use mktg-x). For full-site crawl of a known domain, firecrawl may be better.
Aicoo-Team/AICOO-Skills/skills/build-memory Skill
35★ repoUse this skill when the user wants to build, refresh, or teach their Aicoo agent memory from their real context — files, repo, git history, docs, or Claude/agent memory. Triggers on: 'build my memory', 'build my agent memory', 'teach my agent about me', 'what should my agent know', 'update my memory', 'refresh memory', 'import my context', 'sync my knowledge', 'make my agent know me', 'ingest my repo'.
Aicoo-Team/AICOO-Skills/skills/connect-local-agent Skill
35★ repoUse this skill when the user wants to make THIS machine's coding agent reachable for live agent-to-agent collaboration — start the Aicoo local-agent bridge so paired peers can send requests to their running Claude Code / Codex, with per-tool owner approval. Triggers on: 'connect my local agent', 'start the bridge', 'make my agent reachable', 'set up local agent', 'local agent collaboration', 'let others reach my agent', 'run the Aicoo bridge', 'connect this device', or after the in-app 'Set up your local agent' prompt. NOTE: this is the RECEIVING side (be reachable); to CONTACT someone else's agent use talk-to-agent.
Aicoo-Team/AICOO-Skills/skills/share-agent Skill
35★ repoUse this skill when the user wants to share their AI agent with someone, generate a shareable link, require sign-in, allow anonymous access, let others talk to their agent, configure write access for guests, or manage existing shared links. Triggers on: 'share link', 'agent link', 'share my agent', 'let them talk to my AI', 'require sign-in', 'anonymous link', 'write access', 'edit access', 'guest permissions', or wanting to create a link for investors, prospects, partners, or anyone else to interact with their AI assistant.
Aicoo-Team/AICOO-Skills/skills/talk-to-agent Skill
35★ repoUse this skill when the user wants to contact another person's Aicoo agent (agent-to-agent RPC), send human inbox messages, request/accept agent access, bridge from share links to friend+agent connections, or chat via public share link (`/a/<token>`). Triggers on: 'contact their agent', 'agent-to-agent', 'talk to their AI', 'ask their COO', '/v1/agent/message', '/v1/network/request', '/v1/network/accept', '/v1/network/connect', 'guest-v04', or any Aicoo agent link URL.
saadshahd/moo.md/hunch/skills/soloism Skill
34★ repoUse when two or more concurrent subagents are running through the solo MCP server and one finishing should immediately drive the next, or when spawned subagents go idle unnoticed, linger as stale processes the human must poll or hand-kill, or leave dependent work stalled behind them.
OutlineDriven/odin-claude-plugin/skills/cascade-dedup Skill
34★ repoStrip duplicate and conflicting directives across the system-prompt cascade family: the canonical baseline, the six output-style embeds, and the two external harness embeds.
OutlineDriven/odin-claude-plugin/skills/enhance Skill
34★ repoAudit and tighten agent/plugin surfaces. Use when the user asks to enhance a plugin config, agent definition, skill, or CLAUDE.md.
OutlineDriven/odin-claude-plugin/skills/handoff Skill
34★ repoSnapshot the current session into a resumable handoff artifact so a cold session, agent, or person continues without replaying context. Use when the user says "write a handoff", "hand off", "snapshot this session", "I'm running low on context", "pause and pack this up for the next session", or "resume this later".
OutlineDriven/odin-claude-plugin/skills/llm-self-loop Skill
34★ repoRestructures human-gated workflows into autonomous LLM loops with file-based outputs. Use when a task needs a button click, dashboard check, or human verdict inside its iteration loop.
OutlineDriven/odin-claude-plugin/skills/memory-clean Skill
34★ repoUse when the user asks to audit memory or find stale or duplicate memories.
OutlineDriven/odin-claude-plugin/skills/memory-sanitize Skill
34★ repoUse when sanitizing memory for sharing, redacting PII, or scanning for credentials.
OutlineDriven/odin-claude-plugin/skills/memory-update Skill
34★ repoUse when the user says save this to memory, or scan this session for memories. For auditing existing memories, use memory-clean.
OutlineDriven/odin-claude-plugin/skills/subagent-driven Skill
34★ repoExecute a multi-task plan. Use when the user says "execute with subagents", or hands you an ordered multi-task plan to run.
OutlineDriven/odin-claude-plugin/skills/worktree Skill
34★ repoSet up isolated git worktrees: create a new branch for fresh work, or attach a worktree to an existing branch/PR/commit.
awrshift/claude-memory-kit Skill
34★Persistent memory for Claude Code agents with an agent-audit-ritual architecture. User only talks; the agent captures, audits, proposes promotions, and writes. Memory lives in layers — a hot cache (MEMORY.md) held under three size caps, per-session handoffs (context/handoffs/), topical knowledge articles (knowledge/concepts/), and canonical rules (.claude/rules/) — plus multi-project isolation via projects/<name>/ and an experiments/ sandbox. /close-session runs the end-of-session audit ritual. Zero external dependencies.
WenyuChiou/agent-collab-skills/skills/agent-debate Skill
34★ repoUse when a consequential decision needs adversarial review, opposing agent arguments, a second opinion via debate, or explicit trade-off analysis before implementation.
WenyuChiou/agent-collab-skills/skills/agent-output-reconciler Skill
34★ repoUse when multiple agents have completed a round and the user asks to reconcile outputs, compare Codex and Gemini, synthesize run results, identify conflicts, or decide what should be retried.
WenyuChiou/agent-collab-skills/skills/agent-plan-act-reflect Skill
34★ repoUse when a task needs single-agent self-correction across multiple iterations — write plan, execute, critique own output, revise plan, re-execute, until convergence or budget exhausted. Different from `agent-debate` (which is 2 agents arguing pro vs con); this is 1 agent looping over its own work.
WenyuChiou/agent-collab-skills/skills/agent-shared-memory Skill
34★ repoUse when the user asks to update shared memory, initialize multi-agent memory, summarize decisions so far, identify open questions, or prepare a fresh session primer.
WenyuChiou/agent-collab-skills/skills/agent-task-splitter Skill
34★ repoUse when the user asks to split a goal across Claude, Codex, or Gemini (gemini requests are rerouted - lane deprecated); plan a multi-agent run; break work into parallel agent tasks; or decompose a large task that needs bounded context handoffs. This is the **generic** multi-agent task splitter — writes `.coord/plan.yml` (a DAG) plus per-agent task files. NOT for research-domain routing that touches `.research/`, `.paper/`, or Zotero/Obsidian/NotebookLM ingest pipelines — for those, use `research-hub-multi-ai` instead (different artifact `.coord/multi_ai_plan.md`, research-hub-aware reconciliation).
benjaminard/fable-skills/skills/plain-handoff Skill
34★ repoFinal-summary discipline for agentic sessions. Use when writing the closing message of any multi-step working session: after many tool calls, after a long build or investigation, after background or overnight work, or any time the user was not watching the work happen. Trigger on every end-of-task summary.
FlyFission/nuclear-grade-context-engineering/skills/recording-what-an-agent-did Skill
33★ repoCaptures an agent run's tool calls, decision points, inputs, outputs, token use, and approval steps as a clear, repeatable record linked into the packet trace and verification record. Use when how the agent got there matters for debugging, auditing, cost review, or defending a release decision. Do not use for a read-only run that changes nothing, or to produce a certified compliance audit trail.
FlyFission/nuclear-grade-context-engineering/skills/stress-testing-agent-changes Skill
33★ repoAttacks your own agent change, tool grant, dependency, model, or release on purpose, across risk types such as prompt injection, gaining extra power, unsafe output, and tool misuse, and records what you tried, what happened, and the leftover risk. Use when a change widens an agent's power, data access, or network reach before release. Do not use for a typo fix with no agent power involved, or to produce a certified penetration test or formal security audit.
Datarobot external agent monitoring
datarobot-oss/datarobot-agent-skills/skills/datarobot-external-agent-monitoring Skill
32★ repoInstrument any external or existing AI agent with OpenTelemetry to send traces, logs, and metrics to DataRobot for monitoring, observability, and governance. Use when the user says "add tracing/observability/monitoring to my agent", wants to instrument an existing agent project in their IDE, or wants to send agent traces, logs, or metrics to DataRobot.
jmxt3/gitscape.ai/.agents/skills/context-engineering Skill
32★ repoOptimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to configure rules files and context for a project.
The same file is in 12 other repositories
jmxt3/gitscape.ai/.agents/skills/gpt-taste Skill
32★ repoElite UX/UI & Advanced GSAP Motion Engineer. Enforces Python-driven true randomization for layout variance, strict AIDA page structure, wide editorial typography (bans 6-line wraps), gapless bento grids, strict GSAP ScrollTriggers (pinning, stacking, scrubbing), inline micro-images, and massive section spacing.
The same file is in 13 other repositories
jmxt3/gitscape.ai/.agents/skills/using-agent-skills Skill
32★ repoDiscovers and invokes agent skills. Use when starting a session or when you need to discover which skill applies to the current task. This is the meta-skill that governs how all other skills are discovered and invoked.
The same file is in 1 other repository
jmxt3/gitscape.ai/cli/.agents/skills/addyosmani-agent-skills Skill
32★ repoSpecialist knowledge of the addyosmani/agent-skills project (Shell, JavaScript). Use when working with or extending agent-skills: understanding its architecture, patterns, and configuration. Not for general Shell, JavaScript questions unrelated to agent-skills.
woodfishhhh/EZ_math_model/skills/ez-math-model/external/paper-orchestra/skills/content-refinement-agent Skill
32★ repoStep 5 of the PaperOrchestra pipeline (arXiv:2604.05018). Iteratively refine drafts/paper.tex by simulating peer review and applying targeted revisions, with strict accept/revert halt rules. Maintains a worklog and snapshots each iteration so revert is real, not symbolic. TRIGGER when the orchestrator delegates Step 5 or when the user asks to "refine the draft", "iterate on the paper", or "run peer review on this paper".
woodfishhhh/EZ_math_model/skills/ez-math-model/external/paper-orchestra/skills/literature-review-agent Skill
32★ repoStep 3 of the PaperOrchestra pipeline (arXiv:2604.05018). Execute the literature search strategy from outline.json — discover candidate papers via web search, verify them through Semantic Scholar (Levenshtein > 70 fuzzy title match, temporal cutoff, dedup by paperId), build a BibTeX file, and draft Introduction + Related Work using ≥90% of the verified pool. Runs in parallel with the plotting-agent. TRIGGER when the orchestrator delegates Step 3 or when the user asks to "find citations for my paper", "draft the related work", or "build the bibliography".
The same file is in 2 other repositories
woodfishhhh/EZ_math_model/skills/ez-math-model/external/paper-orchestra/skills/outline-agent Skill
32★ repoStep 1 of the PaperOrchestra pipeline (arXiv:2604.05018). Convert (idea.md, experimental_log.md, template.tex, conference_guidelines.md) into a strict JSON outline containing a plotting plan, literature search plan (Intro + Related Work), and section-level writing plan with citation hints. TRIGGER when the orchestrator delegates Step 1 or when the user asks to "outline a paper from raw materials" or "generate the paper structure".
The same file is in 2 other repositories
woodfishhhh/EZ_math_model/skills/ez-math-model/external/paper-orchestra/skills/plotting-agent Skill
32★ repoStep 2 of the PaperOrchestra pipeline (arXiv:2604.05018). Execute the visualization plan from outline.json — render plots and conceptual diagrams from experimental_log.md and idea.md, optionally refine via VLM critique loop, and produce context-aware captions. Runs in parallel with the literature-review-agent. TRIGGER when the orchestrator delegates Step 2 or when the user asks to "generate the figures for my paper" or "render the plots from this experiment log".
The same file is in 3 other repositories
woodfishhhh/EZ_math_model/skills/ez-math-model/external/paper-orchestra/skills/section-writing-agent Skill
32★ repoStep 4 of the PaperOrchestra pipeline (arXiv:2604.05018). ONE single multimodal LLM call that drafts the remaining paper sections (Abstract, Methodology, Experiments, Conclusion), extracts numeric values from experimental_log.md into LaTeX booktabs tables, splices the generated figures from Step 2, and merges everything into the template that already contains Intro + Related Work from Step 3. TRIGGER when the orchestrator delegates Step 4 or when the user asks to "write the methodology and experiments sections" or "fill in the rest of the paper".
The same file is in 3 other repositories
woodfishhhh/EZ_math_model/skills/ez-math-model/tools/dispatching-parallel-agents Skill
32★ repoUse when EZ_math_model runs in multi or hybrid mode and has two or more independent subtasks that can be assigned to subagents without shared files, shared state, or dataflow dependencies.
woodfishhhh/EZ_math_model/skills/ez-math-model/tools/subagent-driven-development Skill
32★ repoUse when developing or extending EZ_math_model itself from a written implementation plan, especially adding algorithms, tool skills, templates, or multi-step repo changes.
iowarp/clio-kit/io.github.iowarp/chronolog-mcp MCP server
32★ repoChronoLog MCP server implementation using Model Context Protocol
iowarp/clio-kit/io.github.iowarp/compression-mcp MCP server
32★ repoCompression MCP server implementation using Model Context Protocol
iowarp/clio-kit/io.github.iowarp/darshan-mcp MCP server
32★ repoDarshan I/O profiler MCP server for analyzing I/O trace files
iowarp/clio-kit/io.github.iowarp/geo-mcp MCP server
32★ repoMCP server for rendering GeoJSON vector layers into map images with basemaps
iowarp/clio-kit/io.github.iowarp/geojson-mcp MCP server
32★ repoMCP server for inspecting, validating, and summarizing GeoJSON documents (stdlib only)
iowarp/clio-kit/io.github.iowarp/hdf5-mcp MCP server
32★ repoHDF5 FastMCP - Scientific Data Access for AI Agents | CLIO Kit MCP Server
iowarp/clio-kit/io.github.iowarp/ndp-mcp MCP server
32★ repoNational Data Platform (NDP) MCP server for searching and discovering datasets across multiple...