chieflatif

Full audit cycle that dispatches audit agents (security, architecture, correctness, test quality, evidence, product drift, flow, system invariants, dependency intelligence, delivery infrastructure, red team, contract validator), applies consensus logic, and produces a composite report with actionable findings. Use when the user says "audit the code", "review everything", "check for security issues", "do a full review", or wants a comprehensive multi-perspective code review.

Autonomous build orchestrator that executes work orders end-to-end. Use when the user says "continue", "keep going", "build the next thing", "resume", "start building", or wants to make progress on the development plan. Dispatches investigator, tester, prompt-engineering, implementation, and documentation agents in sequence with two-layer quality enforcement (Layer 1 gates + Layer 2 audit agents) and error recovery, while checking for product drift and stale high-impact decisions.

VibeOS audit flow for Codex. Use when the user asks for a code audit, quality review, governance review, architecture review, security review, or wants a full audit cycle before or after implementation work.

Full autonomous-session override for VibeOS in Codex. Use when the user says go autonomous, stop checking in, run on your own, stay in autonomous mode, or wants VibeOS to keep building until a real blocker or decision appears.

Autonomous VibeOS work-order execution for Codex. Use when the user says continue, resume, keep going, build the next thing, start implementing, or wants progress against the current VibeOS plan.

Phase-boundary checkpoint and ratchet review for VibeOS in Codex. Use when the user says phase is done, milestone check, checkpoint this phase, or wants a boundary audit before starting the next phase.

Compact enterprise MVP mission intake for VibOS Comp. Use for competition-grade builds, enterprise design-partner MVPs, rapid production-quality prototypes, battle harness runs, or explicit VibOS Comp/VibeOS Comp requests.

Explain VibeOS concepts and route the user to the right Codex workflow. Use when the user asks what a VibeOS term means, wants the difference between status and project status, needs help choosing a workflow, or wants the system explained in plain English.

Fires on any operation that reads or writes deal, company, contact, call note, or task data. Provides: file schemas, write protocol (entity + index + cross-references + ops.log), account name resolution (fuzzy matching), read-time validation, duplicate detection rules. Output contract: every data write follows the standard protocol and produces a FILES UPDATED block.

Fires when a user asks about deal strategy, how to approach a stalled deal, or needs a business case. Produces: Five-Lens Prism analysis (stakeholder psychology, political capital, competitive dynamics, hidden leverage, temporal dynamics), Three Paths recommendation (velocity/diagnostic/protective), 5-Component POV model. Output contract: three distinct strategic options with evidence grading, never a single recommendation.

Fires on meeting prep, outreach, research, deck creation, and business case building. Provides: company overview, product portfolio, market position, differentiators, pricing context, case studies by vertical, key selling metrics, voice & language patterns, pain signal mapping, and prism insights. Output contract: company-specific context that makes outputs relevant to the prospect's actual situation. When a Platform Document exists (from /research), loads enriched voice, pain, and strategic intelligence.

Fires when a competitor is mentioned, a competitive situation arises, or the user asks how to beat/displace/win against a specific vendor. Produces: per-competitor displacement playbook with strengths, weaknesses, talk tracks, proof points, trap questions, and landmine responses. Triggers on: handle-competition, think-about-deal (competitive lens), battle analysis.

Fires on any written output — emails, outreach, LinkedIn posts, business cases, follow-ups. Provides: voice rules, email formatting, vocabulary substitutions, banned phrases, sign-off, and 7-point voice check. Output contract: all written content matches the user's calibrated tone and style.

Fires when a call transcript is pasted or the user says they finished a call. Produces: 6 structured outputs — MEDDPICC scoring with evidence, task extraction with owners and dates, coaching feedback on 5 dimensions, follow-up email in 3 tones, risk signals with mitigation actions, competitive intel. Output contract: every output is actionable, not just analytical.

Fires when the user requests a presentation AND the Gamma connector is available. Provides: Gamma-specific prompt templates, slide structure guidance, and decision logic for Gamma vs native PPTX path. Output contract: a web-based presentation via Gamma API, or falls back to PPTX if Gamma is unavailable.

Fires on call processing, deal strategy, meeting prep, pipeline review, and deal creation. Provides: 8-element scoring rubric (Metrics, Economic Buyer, Decision Criteria, Decision Process, Paper Process, Identify Pain, Champion, Competition) with Red/Yellow/Green definitions, evidence requirements per level, and discovery questions to advance each element. Output contract: every MEDDPICC score has a specific evidence quote or gap identification.

Use when the founder needs a reusable business artifact from saved workspace context, especially investor/stakeholder updates or competitive snapshots. Combines recent context, tasks, research, closeouts, weekly plans, metrics supplied by the user, and competitor inputs into paste-ready drafts.

Fires when the user pastes a transcript or says they just finished a meeting. Provides: summary, what matters now, risks or blockers, tasks, and follow-up email. Output contract: every closeout leaves the user with a saved record and clear next actions.

Use when the founder needs to identify, connect, verify, or document email, calendar, Drive, CRM, call recording, Apollo, Slack, Teams, search/Tavily, or other operating connectors for SalesSidekick Lite.

Fires when the user wants to get ready for a meeting. Provides: account context, person point of view, talk track angles, questions to ask, watch-outs, and desired next step. Output contract: the user walks into the meeting with a clear point of view and practical talking points.

Use when the user wants to build an outbound SDR campaign, create a target account list, research a market from ICP, find buyers, build account/persona POVs, or draft a 3-5 step outbound sequence. Runs a staged campaign workflow: ICP/value prop/offer intake, 3x long-list, preliminary scan, top-50% deep account research, final prioritized list, top-20 company POVs, buyer persona research, persona POVs, offer mapping, then message sequence.

Fires during setup and daily briefing. Provides: seller identity, company/product context, competitor context, communication style, daily briefing preference, and setup-state logic. Output contract: the system always knows what has been captured and what is still missing.

Fires when the user asks to research a company or person. Provides: company snapshot, person snapshot, likely pressures, point of view for the company, point of view for the person, talk track angles, and questions to ask. Output contract: the result is directly usable for prep, outreach, or the next meeting.

Use when the founder wants to configure Slack channels, extract actionable Slack intelligence, create a daily Slack digest, review Slack decisions and handoffs, or turn selected Slack context into task candidates for the daily operating workflow.

Full audit cycle that dispatches audit agents (security, architecture, correctness, test quality, evidence, product drift, flow, system invariants, dependency intelligence, delivery infrastructure, red team, contract validator), applies consensus logic, and produces a composite report with actionable findings. Use when the user says "audit the code", "review everything", "check for security issues", "do a full review", or wants a comprehensive multi-perspective code review.

Autonomous build orchestrator that executes work orders end-to-end. Use when the user says "continue", "keep going", "build the next thing", "resume", "start building", or wants to make progress on the development plan. Dispatches investigator, tester, prompt-engineering, implementation, and documentation agents in sequence with two-layer quality enforcement (Layer 1 gates + Layer 2 audit agents) and error recovery, while checking for product drift and stale high-impact decisions.