secondme project constitution v0

role: canonical doctrine

use this page when you need the stable answer to:

• what secondme is
• who it is for
• what the first wedge is
• what principles and non-goals guide decisions

if this page conflicts with a dashboard or a snapshot page, this page wins.

one-screen summary

• secondme is a private chief-of-staff operating layer for high-context principals.
• the core bottleneck is not lack of intelligence; it is fragmented context, weak compression, bad timing, and broken follow-through.
• the first wedge is read-only onboarding through one high-context source, starting with telegram.
• the first value is time to first insight; the first wow is tailored leverage already in motion.
• the first user filter is not hnw alone; it is complexity x stakes x agency.
• the moat thesis is not raw model novelty; it is harness quality, memory quality, trust tuning, and continuity over time.
• we are not a research lab; we productize proven patterns, tools, and harnesses that work now.

mission, north star, horizon

mission

preserve and extend continuity of agency for high-context principals.

north star

the user feels:

• i do not need to reconstruct my situation from scratch
• the system found the next best intervention before i asked
• it can act, but it knows when to stop

horizon

• short: win one workflow so strongly that users pull for it repeatedly
• medium: become the private operating layer across communications, planning, memory, and approvals
• long: become the chief-of-staff substrate around a principal, where human and ai operators inherit continuity instead of rebuilding it

why now

the timing claim is simple:

architecture around the model is improving faster than most products are adapting to it.

that matters because:

• harness quality can create order-of-magnitude gains over naked model use on hard tasks
• memory systems and context pipelines are now composable enough for product use
• raw model access is commoditizing, which increases the value of workflow shape, trust, and continuity

weak point:

• benchmark wins do not automatically transfer into our workflow; they must be replicated on principal tasks

who this is for

true filter

• complexity
• stakes
• agency

likely early users

• founders with multiple live fronts
• investors and portfolio operators
• wealth-adjacent principals with fragmented high-stakes context
• operators who already feel the cost of dropped continuity

core pain

the recurring failure mode is:

1. context lives across chats, docs, notes, calendars, tasks, and people
2. the principal repeatedly reconstructs the same situation by hand
3. timing slips and follow-through degrades
4. delegated work loses context and identity
5. assistants sound helpful but fail the trust test

why current agent tooling is not enough

• it still feels session-shaped and operator-heavy
• memory continuity is weak unless someone keeps curating it
• cross-context synthesis exists as craft, not as a stable product loop
• maintenance burden is still high enough to kill adoption

market wedge, first value, first wow, v1

market wedge

read-only onboarding through one high-context source, starting with telegram, for one principal.

first value

time to first insight: the system quickly reflects the user's real situation back from live context with near-zero setup.

first wow

the system explains the user's world back to them, shows where it can create leverage in their exact contexts, and already has 1-2 bounded workstreams in motion.

the user should receive:

• what the system clearly sees about the user's active reality
• where time, attention, or follow-through are leaking
• what long-range goals and bottlenecks are visible
• how the user should work with the system
• what safe workflows, tools, or artifacts are already being prepared

v1 scope

the minimum system to produce that outcome likely includes:

• read-only onboarding from 1 high-context source first, expanding to 2-3 as trust grows
• inspectable memory with promoted summaries
• one principal profile with priorities, preferences, and active campaigns
• self-brief plus capability briefing
• at least one bounded safe workstream already in motion
• briefing generation on a repeatable cadence once the first source is understood
• approval boundary for all meaningful outbound actions
• prepared artifacts such as a relationship map, crm seed, open-loop tracker, draft, or recommendation
• outcome logging so the next cycle starts from updated reality

product invariants

• the harness is the product
• scaffolding > model
• chief of staff > chatbot
• inspectable memory > opaque magic
• continuity quality > response quality
• trust is part of utility
• proactivity is required
• one correct intervention > twenty plausible suggestions
• small adaptive wins > giant demos
• self-improvement must happen through observed outcomes, not fantasy self-critique
• build from proven working components before inventing net-new science

design principles

product

• start with one painful workflow, not category breadth
• optimize for approval-ready action, not insight theater
• reduce maintenance overhead aggressively
• earn more agency through repeated small wins

system

• prefer proven architectures over frontier novelty by default
• use existing strong harnesses and patterns as building blocks
• memory should be inspectable, layered, and promotable
• orchestration should be policy-aware, not just state-aware
• permissions and blast radius shape architecture from day 1

operating

• doctrine is slow; bets are fast
• evidence outranks eloquence
• every bet needs a next test and a kill condition
• raw live channels do not update doctrine directly

self-improvement loop

the system should improve by closing loops on real outcomes:

1. propose a bounded intervention
2. route through the right approval boundary
3. observe what the user accepted, rejected, or edited
4. write back preference, outcome, and confidence signals
5. adjust future action selection, not just future wording

trust and agency ladder

day 1

• ingest
• summarize
• prepare drafts
• recommend next actions

after proof

• schedule or execute reversible low-risk actions with approval
• operate on scoped channels with strong audit trail

after deep trust

• higher-frequency delegation
• broader channel coverage
• narrower approval loops for pre-cleared action classes

business model posture

the business model should follow the trust ladder:

• low-friction entry through obvious proof of power
• paid layer for security, reliability, support, and managed setup
• increasing value through deeper continuity, not feature count

weak point:

• pricing evidence is still weak

proven harness map

inspiration map

active bets

• context management is the core product, not one feature among many
• the first wow is time to first insight plus initiated leverage, not briefing alone
• the wedge is complexity x stakes x agency, not hnw alone
• trust requires a ladder, not one deployment mode
• harness quality matters more than raw model delta for this workflow

non-goals

• being a general research lab
• building a generic multi-agent platform before one sharp workflow wins
• chasing model novelty as a strategy
• autonomy without clear permissions
• category language inflation without new field evidence

doctrine update rule

this document should change only when repeated field evidence, promoted sources, or frozen workflow decisions actually change the map.