Misraj AI — GTM Ecosystem Report

Expansion (upsell/cross-sell) — the third motion Done, task #48/#50

• A crm.lead-only Odoo reader pulls closed-won deals and reconciles them into accounts (migration 058 widens relationship_role to prospect|customer|partner|both and adds customer_deals/account_products/expansion_opportunities).
• The expansion_opportunity agent reasons over cross-sell adjacency against the other 8 products — read-only, capped at 3 candidates per run, with government-adjacent accounts forced onto a manual track twice over (prompt and projection).
• A weekly weeklyExpansionScan job + the new /expansion dashboard (board + drawer) surface it to the operator. Output is recommendations only — no autonomous send anywhere in this surface.
• This closes what the prior milestone below still listed as "Planned, not built" — existing relationships like MNGHA, KAEC, Sukuk Capital, Deloitte, and Libya Diwan can now be worked as expansion once an operator makes the un-park call.

ICP Map — an editable, agent-steering control surface Done, tasks #51 + #81

• The new /icp dashboard page lets an operator edit each sellable product's personas, wedge line, priority rank, status, signal queries, and booking URL — reusing the existing products catalog and account_product_fits, no new tables.
• Critically, those edits now reach the running agents. Before this wave's follow-up fix, dailyAccountSourcing's fit matrix and dailySignalsScan's per-product rotation read the compile-time product catalog, not the DB row an operator had just edited — so an ICP Map change was invisible to the pipeline. Sourcing, signals, and enrollment (including the Worker's backstop) now all read the DB, with a same-payload fallback to the compile-time catalog only when the DB has zero sellable rows.
• The daily contact-enrollment draft also now prefers the product's DB-configured booking URL, so an ICP Map edit to the booking link is live end to end, not just on the page itself.

Dashboard/UI polish + Market Analysis filter Done, tasks #60–#63, #79

• Product filters and fit badges now surface on the Journey Board, Accounts, Contacts, and Signals pages (#60); Market Analysis gained a company-level Misraj vs. the competitor landscape section (#61) plus a per-product filter that scopes the competitor watch and signal feed while keeping the company-wide view always visible (#79, migration 059 tags market_intel_signals by product).
• The Signals Hub is now paginated (12/page) with a dedicated Archived tab (#62, migration 057), and a daily Signal Archivist job auto-archives stale/cold signals so the feed stays current without manual cleanup (#63).

Copy audit + per-product email routing Done, tasks #80, #84

• A system-wide copy pass rebranded 81 strings from Baseer-only framing to the Misraj ecosystem, and renamed the visible "Baseer fit" label to "Misraj fit" — the label operators actually see on every account card.
• Non-Baseer products (Kawn, Seamless Enterprise, Workforces, and the rest) now send sales email from misrajtech.com through a second, dedicated Resend account, with the signature booking link pointed at misrajtech.com/book. Baseer is unchanged — trybaseerocr.com, its own Resend account, trybaseerocr.com/book. The sender resolver fails closed (never silently falls back to the wrong account) if a domain's key is missing.
• Since misrajtech.com's inbound mail isn't live yet, its sends carry an interim Reply-To the verified trybaseerocr.com inbox so replies are never lost.

The multi-product ecosystem (P1–P7) Done

• A products catalog (9 sellable products — بصير/Baseer, Seamless Enterprise, كَون/Kawn, Workforces, Nova-star, 0plus, Compliance RAG, custom app development, Ark) and a new account_product_fits table replace the old single-product assumption; a DB trigger keeps the existing accounts.icp_score reconciled to max(fit_score), so nothing that already read icp_score had to change.
• Proven in prod: account_product_fits holds 288 rows across all 9 products (up from 264 all-baseer). Elm, NHC, and Tawuniya each carry a live 9-product fit matrix with business-specific rationales (e.g. Tawuniya → Compliance RAG scored 62 for SAMA regulation; Elm → Nova-star scored 70).
• Sourcing, contact targeting, the copywriter, the orchestrator's review checks, and funnel reporting are all now product-aware end to end: a candidate account is scored against the whole catalog, its contacts are targeted with that product's own personas, its outreach draft leads with that product's own wedge (not a generic Baseer pitch), and the funnel can be read broken out by product.
• The last open phase — per-product signals/market-intel query rotation (P6) — closed this wave (commits 5842c5e/583eff2): the daily signals scan now rotates its Twitter/news/ LinkedIn queries across the sellable catalog instead of scanning only Baseer/OCR terms every day, and every surfaced signal is tagged with the product it was found under.
• Result: all seven phases of the multi-product plan (P1–P7) are now done and deployed.

Partnerships / BD — a second GTM motion Done, task #47

• A new BD/Partnerships agent sources exploratory partnership opportunities across all 10 partner types (reseller, technology, systems-integrator, and more) — deliberately exploratory-only, with no pricing or commercial terms drafted by the AI.
• Accounts can now be tagged customer, partner, or both; a weekly sourcing job runs every Sunday 07:15 Riyadh; a new Partnerships tab on the dashboard gives the operator a read/curate board.
• Fully human-gated — there is no send or draft action anywhere in this surface; a government-program partner candidate stays on a manual track even after approval (no contacts enrolled, channel forced to manual).

Expansion — the third motion Done — see the 2026-07-10 (late) update above

• This card is kept as the historical record from the 2026-07-09/10 milestone below, written while Expansion was still planned. It has since shipped in full — see "Expansion (upsell/cross-sell) — the third motion" in the milestone update above this one for the built version.
• The point of this motion remains: accounts like MNGHA, KAEC, Sukuk Capital, Deloitte, and Libya Diwan are existing relationships, not cold prospects, and should be approached as expansion now that it's live. MNGHA specifically is a 3-product customer already, but was auto-parked with the government-account exclusion — it still needs an explicit un-park + "existing customer" tag, pending an owner decision (the engine shipping doesn't auto-resolve this operator call).

Proposal generator + Vault versioning Done

• The deck generator now spans the full ecosystem with fit-driven product selection — a capped 3–6 product cards per deck instead of always rendering the whole catalog — plus an optional commercial slide and a mandatory "client need" slide.
• Fixed a single-language bug where an Arabic deck's RTL styling was contaminated by English-only override CSS.
• A new owner-only force-redeliver route regenerates and re-pushes a corrected deck to its existing Vault link without touching Odoo/email/LinkedIn.
• The separate Vault app (report-platform, a different repo, not touched by this wave) now versions every deck edit at the same URL — verified live on the Aramco deck (a re-upload created version 2 at the same link).

Contact-supply recovery Done, root-caused

• Root cause found: the runner was calling Apollo's now-deprecated search endpoints (HTTP 422), silently starving every account-based contact-discovery of Apollo data since launch — the entire prior contact base had been migration imports, not pipeline-sourced. Fixed to call Apollo's current REST API directly.
• Unblocked the orchestrator review queue (a signals-table upsert bug and an over-tight relay timeout were head-blocking every discovery/signal/market-intel output behind one stuck item); added an LLM-relay priority queue so contact-supply work jumps a backlogged relay instead of waiting in line.
• Made the email-first pivot mandatory (not advisory) when a LinkedIn-dormant executive fails the activity gate, and loosened the reveal-corroboration gate so a verified corporate-domain email counts as sufficient on its own — raising yield from the newly-fixed Apollo path.

misrajtech.com + infra/UX Done

• misrajtech.com now redirects to misraj.sa, with its own neutral /book Cal.com link for every non-Baseer product conversation (Baseer keeps trybaseerocr.com/book).
• Outreach emails now render as real HTML with hyperlinked calls to action; the Approvals dashboard preview is WYSIWYG (the exact rendered body + real sender signature); Approvals and Journey pages now cross-link.
• The LinkedIn booking link is now gated to post-engagement only (a reply, a positive signal) — never offered on a first-touch or bare-acceptance message.

Data & scale (as documented)

• ~260 accounts / 225 contacts loaded (V2 migration + old-prospect files), plus new net-new rows landing daily since the sourcing→contacts projection was fixed; 23 government accounts parked (excluded from automated discovery — Apollo has near-zero Saudi-gov coverage).
• 288 account×product fit rows across all 9 products; 87 PR contacts (32 individuals + 55 outlets); 7 competitors tracked.
• Schema at migration 059 as of the 2026-07-10 (late) wave (up from 055); ~1,458+ automated tests passing as of the last documented full run (2026-07-08) — an exact current count was not re-verified in this or the following docs-only pass.

What's next

• Owner decision on un-parking MNGHA (and similar existing relationships) as "existing customer" — the Expansion engine (task #48/#50) is now live, but this specific tagging call is still pending.
• Bring the misrajtech.com inbound MX record online to retire the interim Reply-To workaround; stamp productSlug on the three older agent-runner.ts task-creation sites that still default to Baseer; build the E2E agent-testing harness (task #64).
• Watch the P4 product-match copy checks and the P6 signal-rotation change through their first live week before treating either as fully steady-state.

Three distinct content voices Corrected

• The content planner now writes in three clearly separate voices: Sara (bilingual Misraj founder), Rawan (Arabic business-development / digital-transformation), and Norah — corrected today to an independent Saudi AI enthusiast, not a Misraj voice.
• Norah's content was regenerated as broad AI interest — trends, tools, papers, and ideas she finds genuinely useful — with no product pitching and no Misraj/Baseer affiliation. This keeps her account authentic rather than a disguised brand channel.

Norah X auto-post Live · fail-closed

• Norah's approved, scheduled posts now publish to X automatically through twitterapi.io — no second browser, no manual copy-paste.
• Gated in order: off unless TWITTER_AUTOPOST_ENABLED is exactly on, a 5 posts/day cap, optional business-hours-only window, operator-approved items only, kill-switch/global-pause respected, and idempotent (never double-posts). Drafts never auto-post.

Signals: "reply as Norah" Live · approval-gated

• On any Twitter signal, the operator can now draft a reply in Norah's voice, edit it, and — after an explicit approval — have it posted as a real reply to that tweet.
• The send clears the same execution gate as every other outbound action (content safety, exact-approval match, reply-target guard, kill-switch, idempotency) and is off unless TWITTER_REPLY_ENABLED is on. Recorded in a dedicated table (migration 047).

Signals Hub: multi-source + recency Fixed

• The scan now pulls Twitter, news, and LinkedIn together — a bug that returned zero news was fixed by parsing the news RSS feeds server-side, so all three streams flow into one board.
• A new rolling recency rule (migration 048, default 7 days) drops stale content at every layer — scan, projection, and the surface — so the system only ever studies, replies to, or retweets recent content, never a months-old post it just happened to discover today.

Quality & data model

• ~1,458 automated tests passing (up from ~1,400), covering the Twitter auto-post and reply-as-Norah paths, the multi-source scan, and the recency gate.
• Database schema is now at migration 048 — the twitter-reply drafts table (047) and the signal recency column (048), both additive and fail-closed.

What's next

• Flip on Norah's auto-post and reply flags once the first approved items are queued, and watch the first live posts and replies land cleanly.
• Tune the 7-day recency window per source if a channel proves faster- or slower-moving than the default.

Contact-supply backbone Rebuilt

• The core gap is closed: an approved sourced account or a discovered contact now lands as a real net-new row (the sourcing → accounts → contacts projection). Before today, approved AI outputs were reviewed and then evaporated — nothing became a contact the outreach engine could act on.
• Contact-discovery is now scheduled daily and its runaway loops are fixed — tasks that used to churn to an 18-minute timeout now converge, with a per-task-type tool-round budget and a valid "found none" result.
• A broken linkedin_profile_activity tool (was passing the wrong identifier) was fixed, so the LinkedIn activity gate that disqualifies dormant profiles works again.

Email-first for 17 top executives Live

• An email-first exec gate now routes LinkedIn-inactive GCC decision-makers to email instead of a dead LinkedIn invite. A deterministic Apollo reveal-and-enroll pass ran on the stuck cohort.
• 11 of 17 executives had a verified email revealed via Apollo and were enrolled on the email channel (e.g. geidea, Bank Albilad, Etihad, Jeel, ALDAR).
• 6 had no Apollo email and were returned to the LinkedIn track — including a Deputy Minister who is now flagged for a manual, human-owned approach rather than automation.

Message-draft funnel Repaired & re-voiced

• Outreach drafts were silently quarantining on a partial output shape — fixed, so a valid draft now reaches review. The orchestrator review path that 500-errored on a re-drafted touch is now an idempotent supersede.
• 9 outreach drafts were regenerated in the new voice and are flowing to the approval queue.
• New copywriter rule: no targeting-rationale openers ("given your role/segment…"), and a brief, recipient-first Misraj intro so a cold contact knows who is reaching out — context that earns the pitch, not a sender-centric headline.

Edit-before-approve + a real learning loop Live

• The operator can now edit an AI draft before approving it directly on the approval dashboard, instead of only approve/reject.
• Every edit is captured as a structured learning signal (original AI text vs. the operator's final text) in a new draft_edit_feedback table — the single highest-leverage input for teaching the copywriter what "good" looks like. This closes a gap the 2026-07-04 audit flagged: the learning loop had no outcome linkage.

LinkedIn content auto-publish Live · safety-envelope

• Sara's and Rawan's approved, scheduled LinkedIn posts now publish to their real personal feeds automatically. LinkedIn has no posting API, so this drives the logged-in browser session through the existing MCP (a small additive create_post tool installed on both seats) — no second browser, no re-login.
• Fail-closed by design: off unless explicitly enabled, and every publish must clear a stack of gates — 1 post/day/account, business hours only (Sun–Thu, 09:00–17:00 Riyadh), operator-approved items only, kill-switch/global-pause respected, idempotent (never double-posts). Drafts are never auto-posted.

Personal-account voice + weekly performance review Live

• The content planner was re-grounded on each person's real voice (read live from their accounts). Sara's and Rawan's posts broadened from Baseer-only to the whole Misraj ecosystem (Arabic-AI thought leadership, not a product ad); Norah's X account is treated as an independent Saudi AI enthusiast — broad AI-interest content, not affiliated with the company (operator correction 2026-07-08).
• A Thursday weekly content-performance review now captures per-post engagement. Twitter/X metrics are read for real; LinkedIn per-post metrics are stored honestly as unavailable via the LinkedIn MCP (it exposes no per-post read) rather than fabricated.

Quality & data model

• ~1,400 automated tests passing (up from ~1,202 yesterday), covering the new projection, enrollment, approval-with-edits, and auto-publish paths.
• Database schema is now at migration 046 — adding draft-edit learning capture (045) and per-post content metrics (046) on top of the approval↔AI-draft linkage (042).

What's next

• Watch the first live auto-published LinkedIn posts land cleanly, then loosen the 1/day cap only after several clean posts with no account warning.
• Feed the captured draft-edits back into the copywriter so the learning loop starts improving drafts, not just recording them.
• Work the email-first exec cohort's first approved sends, and progress the manual Deputy-Minister track.

LinkedIn outreach Live

• Connection requests now send from Sara's and Rawan's own LinkedIn seats, capped at 20 invites/day and 100/week per seat, with ban-safety pacing (3–8 minute randomized gaps between invites, max 8 per batch).
• The kill-switch blocking all LinkedIn sends was deliberately turned off today, with an audited database record of the change and who made it.
• First live invites are scheduled from 09:00 Riyadh time today, 2026-07-06.

Inbound CRM pipeline Live

• New CRM leads in Odoo are now automatically researched by an AI agent, turned into a branded proposal deck, and posted back to the CRM as an internal note for sales — no manual work.
• First lead through the full pipeline: Saudi Aramco, with a live, ready-to-share proposal deck link generated automatically.
• Nothing is emailed to the lead directly — the only external action is an internal CRM note visible to the sales team.

Approve-to-send funnel Repaired

• Found and fixed today: approvals were not actually arming anything to send — a broken link between "you approved this" and "the system sends this" meant zero approved messages had ever gone out. That link is now rebuilt and verified end-to-end.
• Every seat now sends from its own identity — sara@trybaseerocr.com and rawan@trybaseerocr.com — with a consistent bilingual sign-off, replacing a shared generic address.

Safety audit Passed

• A full audit of every live flow — LinkedIn, email, inbound, PR, and the agent runtime itself — ran today, producing a 16-fix hardening wave shipped the same day (pacing bugs, stuck-task recovery, duplicate-send protection, and more).
• 1,202 automated tests passing (up from roughly 1,072 a week ago), plus new process-supervision watchdogs that keep the agent runner, scheduler, and LinkedIn bridge alive and self-restarting.

The outbound sales engine (Phase 1)

• Sourcing — discovers GCC large enterprises in document-heavy industries (banking, insurance, government, telecom, healthcare).
• Qualification — a deep ICP rubric scores every account/contact and disqualifies SMEs, off-segment, and non-GCC targets automatically.
• Journey — per-contact state machine: LinkedIn invite → acceptance → your-approved message → 3-business-day reminders → email hand-off.
• Booking — Cal.com bookings close the loop and stop all outreach to that contact.

The demand & brand engine (Phase 2)

• Signals Hub — Twitter/X, LinkedIn, GitHub, community and news signals in one actionable stream.
• Market intelligence — 7 competitors under watch (ABBYY, Nanonets, Upstage…), research papers, tenders, news.
• Content calendar — weekly planning + publish-on-approval for Sara & Rawan (LinkedIn) and Norah (X).
• PR workspace — 20 seeded GCC press contacts, pitch drafting, approval, reply tracking.

The two runtimes

• Hermes agent runtime — runs the LLM reasoning for every agent on the shared Hermes/OAuth configuration (no per-agent API keys). A 1-minute relay carries each agent's request to the model and back.
• Cloudflare edge — the API worker (business logic, approvals, execution, a 5-minute maintenance cron) and the dashboard, both close to GCC users.

The data source of truth — Neon

• 48 database migrations define accounts, contacts, journeys, approvals, execution queue, signals, competitors, content plans, PR, draft-edit learning capture, content-post metrics, twitter-reply drafts, signal recency, and audit tables.
• Every agent output, review, approval, send, and state change is recorded — the dashboard is a live window onto Neon, never a guess.
• Typed status constraints and idempotency keys make the pipeline crash-safe.

Agent ↔ Agent (via the Orchestrator)

Agents never call each other. A sourcing agent's approved accounts become child tasks the Orchestrator spawns for contact-discovery; approved contacts spawn enrichment; enriched contacts spawn outreach drafts. Depth is capped (≤3), duplicates are deduped, and only whitelisted task types can be spawned — so no runaway loops.

Agent ↔ Neon

The runtime claims a queued task from Neon, gives the agent its data + tools, and records the typed output back to Neon — where it is schema-validated. A valid output moves the task to awaiting orchestrator review; an invalid one is quarantined. Heartbeats and a stale-task reaper keep the 24/7 loop crash-safe.

Agent ↔ Dashboard (you)

When an agent proposes something that sends — a message, reminder, tweet, or pitch — it becomes an approval item you see on the dashboard. You approve, reject, or request an edit. Only your approval, matched exactly to the drafted content, lets the execution layer send it. The dashboard also shows agent health, the task board, quarantine, relay status, and the full contact journey for every prospect.

The four prospecting sources

• Apollo — primary structured account + people search.
• Web search — open-web discovery of companies and Vision-2030/digitization signals.
• LinkedIn MCP — GCC contacts, gated on genuine profile activity (dormant profiles disqualified).
• ScrapeGraphAI — deep, Arabic-aware extraction of evidence from a candidate's own web presence.

Every finding, from any source, must clear the ICP rubric before it enters the pipeline.