{"slug": "agent-task-briefing-checklist-10-anti-patterns-free", "title": "Agent task briefing checklist — 10 anti-patterns (free)", "summary": "A developer published a checklist of 10 anti-patterns that cause AI agent tasks to fail, arguing that most breakdowns trace back to the task brief rather than the model. The list pairs each failure mode — such as vague goals like \"make it better,\" unbounded repo write access, and pasted API keys — with a one-line fix, including naming the artifact and acceptance test, allowlisting directories, and pointing to a secret manager instead of embedding credentials. The checklist is offered as a free excerpt from a set of paid briefing templates.", "body_md": "If you hand work to an AI agent (Cursor, Claude Code, Codex, custom agents), most failures trace back to the brief, not the model. Here are 10 anti-patterns I keep seeing, each with a one-line fix. Copy freely.\n\nSmell these in your briefs. Each wastes tokens, time, or trust.\n\n“Make it better.” / “Clean this up.” / “Explore options.”\n\n**Fix:** Name the artifact and the acceptance test.\n\nTools say “repo write: Y” with no path limits.\n\n**Fix:** Allowlist directories; forbid `.env`, secrets, infra by default.\n\n“Research the market” with no table, memo, or decision.\n\n**Fix:** Definition of Done = files a human will open in a meeting.\n\n“Be exhaustive” + “max 10 minutes” + “cite everything.”\n\n**Fix:** Rank constraints; mark Hard vs Soft.\n\nAPI keys in the brief “for convenience.”\n\n**Fix:** Point to a secret manager / env name; never paste values.\n\n“Whatever you need to get it done.”\n\n**Fix:** Escalation table; stop conditions.\n\nNo human owner → agent invents product policy.\n\n**Fix:** Owner field required; acceptance is a human step.\n\n“Write in our voice” with no voice doc or samples.\n\n**Fix:** Link 2 good examples or a 10-line voice sheet.\n\n“Confidence must be 95%.”\n\n**Fix:** Ask for source grades and `assumption` tags instead.\n\nOut of scope empty → agent expands forever.\n\n**Fix:** At least two explicit non-goals.\n\n*This is a free excerpt from a small set of briefing templates I sell. If you want the full fill-in-the-blank packs (briefing, eval, digest, research, incident), they're at [https://briefstack.surge.sh/](https://briefstack.surge.sh/) — the list above is useful on its own either way.*", "url": "https://wpnews.pro/news/agent-task-briefing-checklist-10-anti-patterns-free", "canonical_source": "https://dev.to/briefstack_alex/agent-task-briefing-checklist-10-anti-patterns-free-20nl", "published_at": "2026-09-30 11:32:44+00:00", "updated_at": "2026-09-30 11:47:42.222555+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools"], "entities": ["Cursor", "Claude Code", "Codex"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/agent-task-briefing-checklist-10-anti-patterns-free", "markdown": "https://wpnews.pro/news/agent-task-briefing-checklist-10-anti-patterns-free.md", "text": "https://wpnews.pro/news/agent-task-briefing-checklist-10-anti-patterns-free.txt", "jsonld": "https://wpnews.pro/news/agent-task-briefing-checklist-10-anti-patterns-free.jsonld"}}