← Outcome Pack

X Reach

Get Superdense

npm i -g @nimrobo/superdense

How to start

Make a folder, start your agent in it, and say: "/outcome-setup set up an X Reach outcome pack to grow my qualified reach on X."

One-time setup

Connectors

  • X API — xurl CLI with OAuth2 user context (required): authenticates as your account to read impression_count and own-tweet metrics.

Before your first run

  • ☐Install xurl and authenticate: run `xurl auth oauth2 --app <your-app>` and confirm you can call `/2/users/me`.
  • ☐Verify impression_count is accessible: fetch a recent own tweet with `tweet.fields=public_metrics` and confirm impression_count is non-null.
  • ☐Define your target audience in goal.md — who specifically you want to reach (e.g. 'AI-native engineers', 'B2B SaaS founders').
  • ☐Record your baseline: fetch your last 10 posts and note average impressions and resonance before your first run.

goal.md

  • North star: impressions per post, measured via X API public_metrics (impression_count). A smaller, right audience tends to beat a larger indifferent one — use your judgment when raw reach trades against relevance for your specific goal.
  • Primary metric: impressions per post from X API. Secondary: resonance = (likes + bookmarks) / 1,000 impressions. Track both — a post with high impressions and low resonance may still be worth it depending on your audience strategy.
  • Guardrails to consider: avoid engagement bait and fabricated metrics; prefer originals rooted in real material (a result, a number, a concrete take). Adjust these based on your own content norms and audience expectations.

run.md lever map

  • topic selection: mine concrete audience pain or a shipped result; diagnostics include 2h impressions, resonance (likes+bookmarks/1k), and profile clicks. Topics grounded in real outcomes tend to outperform abstract takes, but test what resonates with your specific audience.
  • hook quality: for replies, consider naming the gap in the parent's own language; for originals, opening with a concrete number or result tends to perform well. Diagnostics: 2h impressions and resonance. Experiment to find what works for your voice.
  • format: original post, reply, or quote-tweet — each behaves differently. Replies tend to drive intent (link clicks); originals and QTs tend to drive reach. QTs without an added link often outperform QTs with one, but this varies by account and topic. Diagnostics: impressions, link clicks per 1k, resonance per format.
  • reply targeting: consider freshness (recent posts still gaining traction) and relevance (outcome or measurement discussions tend to convert better than launch announcements). Prioritize by impression velocity. Diagnostics: link clicks per 1k impressions at 24h.

gate.md

Requires final copy present and human-reviewed, source tweet URL cited for replies and QTs, published post URL recorded in work.md, no fabricated links or metrics. Warning if 24h impression and resonance snapshot is not scheduled after posting.