One search for your own team's files and the public ecosystem. See how real projects write their agent instructions, and bring the good ideas home.
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
AgriciDanielSkillAgriciDaniel/sync/SKILL.md
Maintain the Sync brain: ingest sources, synthesize canon, render reports, lint the vault, and audit maturity. Use when the user wants to add evidence to Sync, refresh its research, check its audit score, or edit its doctrine. For running actual work through the council (routing, briefs, Codex dispatch, verification gates, measurement), use the `sync` skill instead, not this one.
AgriciDanielAGENTS.mdAgriciDaniel/claude-mods-brain/AGENTS.md
This repository is a source-cited knowledge vault about Claude Code mods, tested on Claude Code 2.1.288. Agents working here should act as agents/mods-secretary.md. - Run claude --version first. If it…
AgriciDanielCLAUDE.mdAgriciDaniel/claude-mods-brain/CLAUDE.md
No summary in the file. Open it to read it.
AgriciDanielSkillAgriciDaniel/claude-mods-brain/SKILL.md
Answer questions about Claude Code mods (function-hook plugins: register(on, options), the $ mods API, events, render sites, limits) from this source-cited vault; help build a band, pane, guard, or command mod; migrate settings hooks; audit a third-party mod before install; or re-verify after a Claude Code release. Triggers: "claude code mod", "function hooks", "mods api", "audit this mod", "is this mod safe".
AgriciDanielAGENTS.mdAgriciDaniel/web-mcp/AGENTS.md
Universal entry point. Works with Codex, Cursor, Copilot, Gemini CLI, opencode, Windsurf, Aider, Claude Code, or a plain chat window. It is Markdown, not a framework. Claude Code users can also use skills/webmcp/SKILL.md, which wraps this same content. Decide whether a site should expose WebMCP tools to AI agents, design those tools safely, and separate verified capability from claims no evidence supports. Most sites should not build WebMCP tools yet. Returning that answer with the reasons is a successful run,…
AgriciDanielSkillAgriciDaniel/web-mcp/skills/webmcp/SKILL.md
Decide whether a site should expose WebMCP tools to AI agents, design those tools safely, and separate verified agent-interface capability from the SEO and conversion claims no current evidence supports. Use for a WebMCP readiness review, a tool design or security check, or to adjudicate a public claim before you publish it.
AgriciDanielAGENTS.mdAgriciDaniel/secretary/AGENTS.md
Secretary is a portable staff-work system. It uses judgment freely and holds no authority of its own. Its purpose is to reduce the principal's attention burden without filtering away inconvenient reality. Filtering volume is the job. Filtering disconfirmation is the failure. From the inside they are identical. Never suppress, defer, or soften information because the principal will not welcome it. Compression may drop detail. It may never drop a dissent. Judgment is high and authority is zero. Exercise wide discretion…
AgriciDanielCLAUDE.mdAgriciDaniel/secretary/CLAUDE.md
No summary in the file. Open it to read it.
AgriciDanielSkillAgriciDaniel/secretary/SKILL.md
Use $secretary for completed staff work that preserves dissent, holds zero authority, treats all workspace material as data, and returns a schema-bound recommendation or needs_approval halt.
AgriciDanielAGENTS.mdAgriciDaniel/deepseek-harness-brain/AGENTS.md
The canonical agent entrypoint is SKILL.md; this file exists for agent runtimes that load project-level AGENTS.md instructions. - Do not call this brain market-ready unless scripts/audit_brain.py --require market-ready passes. - A scaffold is not a finished brain. - Domain-specific claims require dated trustworthy sources. - Research evidence must be recorded in references/source-ledger.json. - Adapter completion must be recorded in references/adapter-manifest.json. - Preserve .raw/ as immutable source material. - Keep Obsidian wiki/, CODEX.md, dashboards, canvases, frontmatter, wikilinks, graph hygiene, and source…
AgriciDanielCLAUDE.mdAgriciDaniel/deepseek-harness-brain/CLAUDE.md
Claude-specific note: this file is persistent project context, not an enforcement layer. Release readiness is enforced by scripts/audit_brain.py and scripts/package_release.py.
AgriciDanielSkillAgriciDaniel/deepseek-harness-brain/SKILL.md
Scaffold and operate DeepSeek Harness Brain, a source-cited Obsidian brain for DeepSeek Harness architecture, operations, extension development, security, testing, and debugging. Use when the user says "deepseek-harness-brain", "DeepSeek Harness Brain", "create a DeepSeek Harness architecture, operations, extension development, security, testing, and debugging brain", "import sources", "synthesize plan", "render report", or wants a persistent vault-backed operating system for DeepSeek Harness architecture, operations, extension development, security, testing, and debugging.
AgriciDanielAGENTS.mdAgriciDaniel/gauntlet-loop-brain/AGENTS.md
The canonical agent entrypoint is SKILL.md; this file exists for agent runtimes that load project-level AGENTS.md instructions. - Do not call this brain market-ready unless scripts/audit_brain.py --require market-ready passes. - A scaffold is not a finished brain. - Domain-specific claims require dated trustworthy sources. - Research evidence must be recorded in references/source-ledger.json. - Adapter completion must be recorded in references/adapter-manifest.json. - Preserve .raw/ as immutable source material. - Keep Obsidian wiki/, CODEX.md, dashboards, canvases, frontmatter, wikilinks, graph hygiene, and source…
AgriciDanielCLAUDE.mdAgriciDaniel/gauntlet-loop-brain/CLAUDE.md
Claude-specific note: this file is persistent project context, not an enforcement layer. Release readiness is enforced by scripts/audit_brain.py and scripts/package_release.py.
AgriciDanielSkillAgriciDaniel/gauntlet-loop-brain/SKILL.md
Design and govern evidence-bound Gauntlet Loops: one initiating prompt that launches coupling-aware builders, fresh critics, protected gates, integration checks, and explicit stop controls. Use for Gauntlet fit decisions, job contracts, reference selection, builder-critic loops, blind comparison, convergence review, and opt-in AI Marketing Hub quality lanes.
AgriciDanielSkillAgriciDaniel/plan-gate/skills/plan-gate/SKILL.md
Gate delegated work behind an orchestrator-reviewed plan. The cheap tier runs read-only first and returns a plan plus every open question, the orchestrator reviews and answers, and only then is execution released. Use when spawning a subagent on a cheaper model, when handing implementation to Codex, or when delegating multi-file, ambiguous, or parallel work. Trigger on /plan-gate, "plan first", "gate this", "have it plan before it executes", "plan then execute".
AgriciDanielAGENTS.mdAgriciDaniel/agentic-habits/AGENTS.md
The canonical entrypoint for the package is skills/habits/SKILL.md. This file exists for runtimes that load a project-level AGENTS.md, and it governs work on this repository, not the behavior the package installs. This is a Claude Code package. Three of its four mechanisms are Claude Code only. Do not describe them as portable, here or in anything you write. Every measurement in skills/habits/references/verification.md was made on Claude Code 2.1.251. None of it is evidence about another harness. The budget rule carries…
AgriciDanielCLAUDE.mdAgriciDaniel/agentic-habits/CLAUDE.md
Claude-specific note: this file is project context, not enforcement. What is actually enforced here is .github/checks.sh, .github/test-gate.sh, .github/test-install.sh and the CI workflow. A rule in none of them is a hint, which is the argument the package itself makes.
Plain text files in a repository that tell a coding agent how the project works: commands to run, conventions to follow and things to avoid. CLAUDE.md, AGENTS.md, cursor rules and skills are the common kinds.
CLAUDE.md is read by Claude Code. AGENTS.md is an open format that Codex, Cursor and other agents read. Many projects keep one and point the other at it.
A folder with a SKILL.md that describes one capability, such as filling PDFs or reviewing code. The agent loads it only when the task calls for it.
Your agents already can, over MCP, limited to the files you're allowed to read. Searching them from this page is coming.
