hoangnguyen0403/agent-skills-standard/common-task-complexity-routing

Task Complexity Routing (SNC)

Priority: P1 (HIGH)

Same ticket label, different engineering difficulty. Score the task, not the label, then route by tier.

Score

Sum the three, never average. Output line: SNC: S=[0-2] N=[0-2] C=[0-2] total=[n] tier=[low|medium|high].

Tiers

• tier=low: total 0-2
• tier=medium: total 3-4
• tier=high: total 5-6

Routing

fast | standard | strong is a capability hint only. Each host maps it to its model ladder and binds the named review roles to its own invocation mechanism; the hint is not authorization: premium spend, account selection and execution mode remain host/human decisions.

Capability and Effort

Capability and thinking effort are independent dimensions. Recommend the least expensive eligible configuration adequate for the task and its verification burden; never simply the cheapest model. The host resolves eligibility, approved profiles, supported effort and authorized fallback.

• Deterministic operation uses an existing tool, script or CLI before any LLM call; bounded implementation uses tier-appropriate capability with focused tests.
• Ambiguous investigation: wider effort, because premature narrowing costs more than the extra reasoning.
• Consequential review: capable model plus an independent reviewer; a small diff never lowers a sensitivity floor.

Explicit configured model and effort settings are preserved, not silently replaced. Unsupported or unknown effort is recorded unsupported/unknown, never mapped across vendors. Resolved settings, receipt fields, cost semantics and recovery routes: Capability and effort resolution.

Sensitive-Change Risk Floor

Sensitive changes touching authentication, authorization, payments/financial data, data integrity, secrets, or system trust boundaries have an explicit risk floor independent of arithmetic SNC totals:

• Floor: Minimum tier=medium (guided execution, deep review, human plan acknowledgment) even if arithmetic total is ≤2 (e.g. S=0, N=0, C=2).
• High Sensitivity: If spread or novelty is non-zero (S≥1 or N≥1 with C=2), enforce tier=high (plan-first HARD STOP, independent security review, human merge approval).
• Irreversible Actions: Actions involving data destruction, schema drops, credential rotation, privilege escalation, or production cutover strictly require explicit human authorization regardless of tier.

Rules

• Score before planning; emit snc_tier and model_tier in every Handoff Payload.
• Record requested capability, observable actual configuration, selection reason, effort support and limitations in the existing receipt; unknown stays unknown.
• Label the score as inference when derived from ticket text alone; re-score after specialist-codebase-scout reports impact radius.
• Round up on doubt. Any auth, money, trust-boundary, or hot-path touch is C=2 and subject to the sensitive-change risk floor.
• Downward complexity reassessment is permitted ONLY with documented new evidence (e.g., scouted blast radius proves isolated scope with verified non-interference), NEVER to bypass unresolved risk, required approvals, or the sensitive-change risk floor.
• Re-score when scope grows mid-task; the tier must rise immediately when expansion is detected.
• Never downgrade a tier to skip a gate. Ticket type (bug vs feature) never sets the tier.

Red Flags

Stop and re-score when hearing or thinking: "it's just a bug fix", "one-line change", "skip the plan, it's small", "we'll review it later", "mark it low so we can merge tonight". Each is a rationalization, not a score.

Anti-Patterns

• No tier by ticket type: "bug" and "feature" say nothing about S, N, or C.
• No averaging: three 1s and one 2/0/1 both route differently only when summed.
• No silent tier: a plan or handoff without snc_tier is incomplete.
• No high tier without named reviewers: tier=high must list which specialists review before merge.
• No premium by tier: a strong hint never authorizes premium spend or a host account change.
• No cheapest-model reflex or silent effort mapping: adequacy includes verification burden and repair cost; vendor effort levels are not equivalent.

References

• SNC Rubric with Worked Examples
• Routing Table Detail