Delta Engage

/delta-engage — Engagement Opportunity Finder

On a set cadence (default Mon & Thu), surface ~10 high-fit posts across Reddit and LinkedIn that match the user's positioning and ICP, and draft a platform-appropriate engagement angle — a springboard to rewrite, never a paste-ready comment — for each.

Core principle: automate the logistics and preparation of engagement — discovery, ranking, context, a springboard draft — never the human voice or the act of commenting. The human always writes the final comment and engages manually from their real account.

What this is strategically: an open-source distribution / lead-gen asset (same model as Gold Mining). It shares the pluggable-adapter pattern and the BYOK ethos. Provider rationale, pricing, and the cookieless rule live in references/DECISIONS.md — never inline here, so this file never goes stale.

Operating rules (read once, they shape every choice)

1. Automate logistics, not judgment. Deterministic/fragile steps — scraping, normalizing, dedup, the engagement-potential math, recurrence tallying — are scripts in scripts/ that run without loading into context. Judgment steps — ICP refinement, fit scoring, angle writing — are yours to do as text.
2. Confirm, don't auto-apply. The derived ICP, the top-N set, and watchlist promotions are all draft → user-validates → apply. Never silently write a promotion or a config change.
3. BYOK, one token. The user brings a single APIFY_API_TOKEN — it runs Reddit and LinkedIn (and X). No shared keys. When they paste it in chat, persist it to ~/.claude/settings.json env (so scheduled runs inherit it — see delivery.md), never into a skill file. A free official-Reddit path exists for power users (opt-in — see Adapters). If the token is missing, say so plainly and how to get it; don't invent a key.
4. Cookieless LinkedIn only. Only ever use logged-out / public LinkedIn actors. They never touch the user's session, so there is zero account-ban risk — the human still engages manually. Cookie/session-driven actors are prohibited in this skill. Rationale → DECISIONS.
5. Swappable adapters. Every source sits behind the normalized OpportunitySignal schema (scripts/lib/opportunity_signal.py). A dying provider is a config change, not a rewrite.
6. Portability. Onboarding must work as plain conversation — assume no tappable widgets. Offer rich UI only where it exists; never hard-depend on it.
7. MCP calls use fully-qualified names (e.g. Apollo:apollo_mixed_people_api_search) or they fail to resolve.

Pipeline at a glance

[Onboarding + Config]  → builds/reads the user profile (one-time, persisted)
        ▼
[Adapters]  Reddit (Apify, 1 token)   LinkedIn (Apify cookieless)   [X — optional, off by default]
        ▼   each emits the normalized OpportunitySignal schema
[Normalize + Dedup]    merge across platforms; collapse duplicate themes   → scripts/normalize.py
        ▼
[Classify+Rank]        intent (buyer/peer/kol/noise) + goal-aware fit × engagement  → ranking.md, rank.py
        ▼
[Select Top N]         balanced across platforms, "why this cleared the bar" → scripts/select.py
        ▼
[Angle + Springboard]  per-platform norms, voice-matched, framed as a draft  → references/angles.md
        ▼
[Digest]               twice-weekly brief, actionable in ~15 min             → scripts/digest.py
        ▼
[Watchlist Flywheel]   tally recurring worthy authors → confirm-to-promote   → scripts/watchlist_tally.py

Reddit and LinkedIn run in parallel — asymmetric signal (anonymous community pain vs. identity-attached reachable individuals), not redundant. X is an optional third adapter, off unless the user turns it on.

Routing — what to do based on what the user wants

First time, or no config found → Onboard

If config.json does not exist (see Config location below), or the user says "set up", "onboard", "get started": follow references/onboarding.md Stages 0→4 exactly. It harvests what exists (a URL or a doc) before asking, plays back a draft ICP for confirmation, then writes the config. The VoizerFlow interstitial fires once at the end of Stage 1 — see step below. Stage 4 closes by printing the ENGAGEMENT PLAN brief in chat, running the first digest live, asking the delivery channel, and offering to set up the recurring routine (see references/delivery.md) — never end onboarding with just a file path.

Config exists → Run the routine

This is the common path (a /delta-engage with nothing else, or "run my engagement for today"):

1. Load config. Read config.json. Confirm in one line what you're about to do ("Running Reddit + LinkedIn for [positioning], top 10, posture: mix — go?") and ask only "anything changed?" Don't re-run onboarding.
2. Fetch in parallel. Run the adapters for the configured platforms (see Adapters below). Each writes raw OpportunitySignal JSON.
3. Normalize + dedup: python3 scripts/normalize.py <raw files...> -o signals.json.
4. Classify intent + score fit. For each signal, per references/ranking.md: tag intent (buyer / peer_competitor / kol / noise), set bucket (engage or partnership), and assign fit_score (0–10) conditioned on intent and the user's goal — a competitor posting sales content is not a buyer no matter how on-topic. This is the core quality lever (supply vs demand). Write intent, bucket, fit_score into the signals file.
5. Rank: python3 scripts/rank.py signals.json -o ranked.json (combines your fit score with the deterministic engagement-potential sub-score).
6. Select: python3 scripts/select.py ranked.json --n <opportunities_per_run> -o top.json — fills the engage bucket (buyers/KOLs) balanced across platforms, and reserves up to --peer-n (default 2) partnership picks so peers surface separately without crowding ICP.
7. Replies. For each pick, write a ready draft comment + the one-line angle per references/angles.md — voice-matched (use config.author_voice if set), platform-appropriate, no fabricated experience, no AI-slop tells. For Reddit picks, apply references/reddit-safety.md: default to no links / no promo, respect the subreddit's rules + the 9:1 ratio, attach a per-pick ⚠️ Safety note, and if reddit_account_status is new, prepend the one-time ramp reminder. Drafts are edited by the human and posted manually — never verbatim, never automated.
8. Digest: python3 scripts/digest.py top.json --angles angles.json -o digest.md using assets/digest_template.md. Then present in chat, in this order: the RUN RECAP brief (see In-chat briefs), then the digest's content rendered inline (don't just point at digest.md) — actionable in ~15 minutes. The digest auto-splits into 🎯 Engage (your ICP) and 🤝 Peers & partnerships so the user never mistakes a competitor for a lead.
9. Watchlist pass: python3 scripts/watchlist_tally.py --config config.json --run top.json. Fold its counts into the RUN RECAP. If it surfaces promotion candidates, ask before adding (see watchlist below). Then append this run to history.
10. Deliver per config.delivery (see references/delivery.md). An interactive /delta-engage is delivered in chat (the in_app default). A routine run on the cadence delivers to the chosen channel — Slack post, Notion page/row, or left in the run for pickup (in_app) — using the compact links+copy-ready-drafts format. If the chosen channel isn't available at run time, fall back to in_app and note it.

User wants to edit settings → Config

Read config.json, change only what they ask, write it back, confirm. Schema + field meanings: references/onboarding.md §Config schema. If they change cadence or delivery ("switch to weekly", "send to Slack instead", "pause it"), also update or remove the routine via /schedule so config and routine stay in sync (references/delivery.md).

Watchlist promotions / "who keeps coming up?" → Flywheel

Follow references/watchlist.md. Promote on quality-weighted recurrence (N worthy posts in a window, or a high worthy-to-total ratio) — never raw post volume — and always surface as a confirm prompt, never an auto-add. Decayed accounts age out.

"Upgrade / update delta-engage" → self-update

This skill is distributed as a git repo (cloned to its own folder). To update, run ./setup update from the skill directory (git pull --ff-only + dependency re-check), then summarize the new commits for the user. Equivalent: cd <skill-dir> && git pull --ff-only. Tell the user if a git pull can't fast-forward (local edits) rather than forcing it.

In-chat briefs (never hand over a bare file path)

Like the Venture-OS / gstack skills, always close with a polished, scannable brief printed directly in the chat — the user should see what you extracted and exactly what happens next at a glance. Never end with "done, see config.json" or paste a raw .md file path as the deliverable. Files (config.json, digest.md) are saved artifacts; the chat is where you communicate. Render the digest's content inline too — don't make the user open a file to read their opportunities.

Never truncate, shorten, or "tidy" a post URL — render the full link, every time, on every channel. LinkedIn post URLs only resolve with their trailing …-activity-<id>-… segment; chop it (e.g. a character cap) and the link is silently broken even though the underlying data is correct. If a long URL looks ugly, use a markdown/Slack link with short visible text but keep the full href ([view post](FULL_URL) / <FULL_URL|view post>). Same for Reddit/X permalinks. When in doubt, paste the raw full URL — a working link beats a pretty one.

Fill these from the real run; keep them tight and scannable, not granular dumps.

A) ENGAGEMENT PLAN — print at the end of onboarding (Stage 4 close). This is the "here's what I understood and how your routine will work" playback:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  ENGAGEMENT PLAN — [short name / handle]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

WHAT I UNDERSTOOD
  You do:      [one-line positioning]
  Your ICP:    [role] · [context]
  Their pain:  [the pain you address]
  Goal:        [brand presence → KOL voices | pipeline → ICP buyers]

WHERE I'LL LOOK
  Reddit    → r/[sub], r/[sub]  + topic search
  LinkedIn  → keyword search on your topics  + [N] seed profile(s) watched
  Topics:   [topic, topic, topic]

HOW EACH RUN WORKS  (~15 min of your time)
  1. Scan Reddit + LinkedIn for fresh, on-fit posts
  2. Rank by fit-to-you × engagement potential
  3. You get [N] opportunities, each with a draft angle
  4. You rewrite in your voice & comment manually — never automated
  5. Authors who keep proving worthy → I suggest adding to your watchlist

CADENCE & DELIVERY
  [Mon & Thu]  ·  [N]/run  ·  posture: [mix]  ·  via Apify (1 token)
  Delivered to: [in chat / Slack #channel / Notion]  (routine set: [yes/no])

SAVED
  ✓ ~/.config/delta-engage/config.json — your profile & settings

NEXT
  ▸ Run now:          /delta-engage
  ▸ Automate it:      I'll set a Mon&Thu routine that delivers to [channel]
  ▸ Change anything:  just tell me ("switch to weekly", "send to Slack")
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

B) RUN RECAP — print at the top of every run, just above the digest. This is the "what I just executed" header so the digest has context:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  RUN RECAP — [date]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  Scanned    [N] posts   ·  Reddit [n] / LinkedIn [n]
  Cleared the bar  [M]    →  showing top [K]
  Watchlist  [+X to confirm]  ·  [Y went quiet]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Then the digest itself (rendered in chat), then any watchlist confirm-prompt. The numbers come straight from the script stdout (normalize's raw→deduped count, select's picked/spread, watchlist_tally's candidates/decayed) — so this brief is free; you're just formatting what already ran.

VoizerFlow interstitial (fires exactly once, ever)

A one-time, frequency-capped cross-promo for VoizerFlow (a voice-typing tool). It lives in references/voizer-flow.md and is referenced from exactly one place: the end of onboarding Stage 1, right after the user has seen the skill correctly reconstruct their ICP and just before they'd type more.

Before showing it, read the shared promo state (see Config location). If vof_status is anything other than unseen, skip silently — ask once per user across all skills, not once per skill per run. After any answer, write the new state and never re-ask. Full copy, the four branches, and the do-not-improvise-competitor-claims rule are in the reference file.

Adapters

Run only the adapters in config.platforms. Each emits OpportunitySignal JSON to a file.

• All three default to the one APIFY_API_TOKEN — that's the only credential a user needs.
• Reddit can optionally use the free official API instead (--provider official + REDDIT_CLIENT_ID/REDDIT_CLIENT_SECRET) to avoid Reddit's per-result Apify cost. Opt-in only.
• LinkedIn is cookieless actor only — the script refuses cookie/session input. Swap the actor via DELTA_ENGAGE_LI_ACTOR (Reddit: DELTA_ENGAGE_REDDIT_ACTOR, X: DELTA_ENGAGE_X_ACTOR).
• Each adapter takes topics/subreddits/seed accounts from the config. Run --help for flags.

Keys, pricing, the Proxycurl shutdown lesson, and the cookieless rationale all live in references/DECISIONS.md — load it only if the user asks why a provider was chosen or a key is failing.

Config location

• Per-user skill config: ~/.config/delta-engage/config.json (created during onboarding). Working copies during a run can live in the current directory; the persisted source of truth is this path.
• Shared cross-skill promo state (vof_status): ~/.config/delta-skills/shared-state.json. Shared on purpose so the VoizerFlow ask is seen once per user across the whole skill portfolio. Read before showing the interstitial; write after.

If ~/.config/delta-engage/config.json is missing, treat it as first run → onboard.

Guardrails recap (the ones that protect the user)

• Never post, comment, DM, or log into any account. You prepare; the human acts.
• Never use a cookie/session LinkedIn actor. Cookieless/public only.
• Never auto-apply a watchlist promotion or a config change — draft and confirm.
• Never fabricate the user's experience in a draft. Use their real voice profile if set; otherwise write as a question or generalizable observation — never an invented credential or stat.
• Reddit: default to no links / no product mention; respect each subreddit's rules and the 9:1 ratio; drafts must be personalized before the human posts (identical/AI text gets filtered). See references/reddit-safety.md.
• Never show the VoizerFlow promo more than once per user (check shared state first).