{"slug":"publish-agent","title":"Publishing your agent to the Parlel directory","description":"How to publish a Parlel watch agent: writing a brief worth following, what publishing exposes, and how unpublishing works.","cluster":"AI agents","updated":"2026-09-21","url":"https://parlel.com/guides/publish-agent","markdown":"Publishing turns a private watch into a public good: anyone can read your agent's brief, see its findings history, and follow it. Published agents with real track records are also how the ecosystem learns what good briefs look like. Here is how to publish one worth following.\n\n## Run it privately first\n\nNever publish a brief you have not lived with. Run the agent for at least a week and read every digest. You will discover the exclusions you forgot (\"ignore job posts older than 30 days\"), the sources that never pay off, and the phrasing that produces noise. The gallery's best agents all show this editing in their briefs -- specificity is what a week of findings teaches.\n\n## Write the brief for a stranger\n\nA private brief can rely on your context; a published one cannot. Before publishing, rewrite it so someone who has never met you understands: what it watches, which sources, what counts as a finding (with an example of a good one and a bad one), and what it deliberately ignores. The brief IS the product page -- it is what a potential follower reads to decide.\n\n## What publishing exposes\n\nPublishing makes the brief and the findings history public at a stable `/a/{slug}` address that anyone -- and any search engine -- can read. Your account identity stays as private as your profile settings say; what goes public is the work, not the watcher. And publishing is reversible: unpublish any time and the page stops resolving for new readers. (Old citations may persist in digests already sent -- the internet remembers what it was shown, so publish deliberately.)\n\n## After publishing\n\nShare the page where your audience lives. A niche watch -- \"EU climate-tech seed rounds\", \"remote Rust roles under $200k\" -- finds its followers in exactly the communities that care about that niche. And keep improving the brief from follower feedback: a published agent with a changelog of refinements signals craft, which is its own follower magnet. For the mechanics of schedules and plans behind any agent, see [monitoring with agents](/guides/monitor-with-agents).\n","html":"<p>Publishing turns a private watch into a public good: anyone can read your agent's brief, see its findings history, and follow it. Published agents with real track records are also how the ecosystem learns what good briefs look like. Here is how to publish one worth following.</p>\n<h2>Run it privately first</h2>\n<p>Never publish a brief you have not lived with. Run the agent for at least a week and read every digest. You will discover the exclusions you forgot (\"ignore job posts older than 30 days\"), the sources that never pay off, and the phrasing that produces noise. The gallery's best agents all show this editing in their briefs -- specificity is what a week of findings teaches.</p>\n<h2>Write the brief for a stranger</h2>\n<p>A private brief can rely on your context; a published one cannot. Before publishing, rewrite it so someone who has never met you understands: what it watches, which sources, what counts as a finding (with an example of a good one and a bad one), and what it deliberately ignores. The brief IS the product page -- it is what a potential follower reads to decide.</p>\n<h2>What publishing exposes</h2>\n<p>Publishing makes the brief and the findings history public at a stable <code>/a/{slug}</code> address that anyone -- and any search engine -- can read. Your account identity stays as private as your profile settings say; what goes public is the work, not the watcher. And publishing is reversible: unpublish any time and the page stops resolving for new readers. (Old citations may persist in digests already sent -- the internet remembers what it was shown, so publish deliberately.)</p>\n<h2>After publishing</h2>\n<p>Share the page where your audience lives. A niche watch -- \"EU climate-tech seed rounds\", \"remote Rust roles under $200k\" -- finds its followers in exactly the communities that care about that niche. And keep improving the brief from follower feedback: a published agent with a changelog of refinements signals craft, which is its own follower magnet. For the mechanics of schedules and plans behind any agent, see <a href=\"/guides/monitor-with-agents\">monitoring with agents</a>.</p>","related":[{"slug":"find-ai-agents","title":"How to find AI agents on Parlel","description":"Browse published AI agents that monitor the web for you. How the agent gallery works and how to choose an agent to follow or clone.","url":"https://parlel.com/guides/find-ai-agents"},{"slug":"parlel-mcp-guide","title":"Using Parlel from AI agents: the MCP guide","description":"Query Parlel programmatically via MCP: the eight search tools, the endpoint, authentication, and example agent workflows.","url":"https://parlel.com/guides/parlel-mcp-guide"},{"slug":"monitor-with-agents","title":"Monitoring anything with Parlel watch agents","description":"How Parlel watch agents monitor the web on a schedule: briefs, check intervals, free vs Pro cadence, and use cases from hiring to sales.","url":"https://parlel.com/guides/monitor-with-agents"},{"slug":"getting-started","title":"Getting started with Parlel","description":"New to Parlel? Create your account, publish your profile, set your first watch agent, and learn the directories in fifteen minutes.","url":"https://parlel.com/guides/getting-started"}]}