ozerohax/project-discovery-mode-selector

<purpose> <item>Choose the right planning mode before any project artifacts are produced</item> <item>Prevent over-planning simple requests and under-planning risky initiatives</item> </purpose>

<when_to_use> <item importance="critical">At the start of project work, before project-fast-init or project-standart-brief</item> <item importance="high">When the request is ambiguous and mode choice can change scope, timeline, and risks</item> </when_to_use>

<required_preload> <item>shared-base-rules</item> <item>shared-docs-paths</item> <item>planning-task-classifier</item> <item>planning-risk-assessment</item> <item>planning-scope-definition</item> </required_preload>

<inputs> <required>User request and expected business outcome</required> <required>Known constraints: timeline, budget, team size, compliance</required> <optional>Technical context: integrations, migrations, public APIs, unknown dependencies</optional> <optional>Change impact signals: number of modules, teams, and affected contracts</optional> </inputs> <method> <step>Classify request complexity: small / medium / high</step> <step>Evaluate risk factors: compliance, migrations, external dependencies, public contracts</step> <step>Estimate uncertainty: count of critical unknowns that block planning</step> <step>Select mode using decision rules and confidence level</step> <step>Record rationale and escalation triggers</step> </method>

<decision_rules> <fast_mode> <condition>Use when complexity is small or medium, risks are controllable, and unknowns are limited</condition> <condition>Typical indicators: one team, low compliance pressure, no critical migration, no public API breakage</condition> <condition>Time pressure is high and a one-iteration plan is acceptable</condition> </fast_mode> <standart_mode> <condition>Use when complexity or risk is high, or unknowns can invalidate architecture decisions</condition> <condition>Typical indicators: multi-team coordination, compliance/security constraints, migrations, external contracts</condition> <condition>Traceability from brief to PRD/arch/epics/tasks is required before implementation</condition> </standart_mode> <escalation> <condition>Start in fast mode, then escalate to standart mode if critical unknowns remain after pulse-scan</condition> <condition>Escalate when gate is CONCERNS/FAIL and corrective actions exceed fast timebox</condition> </escalation> </decision_rules>

<output_format>

<section>Chosen mode: fast | standart</section> <section>Confidence: high | medium | low</section> <section>Top reasons (3-5 bullets)</section> <section>Primary risks and constraints</section> <section>Escalation triggers</section> <section>Next step command (exactly one): `project-fast-init` or `project-standart-brief`</section> </output_format>

<quality_rules> <rule importance="critical">Mode decision must be justified by explicit risk/complexity criteria, not preference</rule> <rule importance="critical">If confidence is low, include concrete data needed to confirm or revise decision</rule> <rule importance="high">Escalation triggers are mandatory for fast mode</rule> </quality_rules>

<do_not> <item importance="critical">Do not start planning artifacts before selecting mode</item> <item importance="high">Do not select fast mode when compliance/migration/public-contract risks are unresolved</item> <item importance="high">Do not select standart mode without evidence of complexity or risk</item> </do_not>