Add Computational Complexity Section to Scope Skill Output Template

[Trecek]

Summary

Add a Computational Complexity section to the scope skill's output template in SKILL.md, positioned between the existing "Technical Context" and "Hypotheses" sections. The section contains 4 fields: Dominant operation, Scaling behavior, Known bottlenecks, and Gotchas. An instruction emphasizing baseline computation costs is added to the "Known bottlenecks" field. A corresponding subagent instruction is added to Step 1 so the scope workflow actively gathers complexity information. A contract test is added to validate the section's presence and field completeness.

Architecture Impact

Process Flow Diagram

%%{init: {'flowchart': {'nodeSpacing': 40, 'rankSpacing': 55, 'curve': 'basis'}}}%%
flowchart TB
    %% CLASS DEFINITIONS %%
    classDef terminal fill:#1a237e,stroke:#7986cb,stroke-width:2px,color:#fff;
    classDef stateNode fill:#004d40,stroke:#4db6ac,stroke-width:2px,color:#fff;
    classDef handler fill:#e65100,stroke:#ffb74d,stroke-width:2px,color:#fff;
    classDef phase fill:#6a1b9a,stroke:#ba68c8,stroke-width:2px,color:#fff;
    classDef newComponent fill:#2e7d32,stroke:#81c784,stroke-width:2px,color:#fff;
    classDef output fill:#00695c,stroke:#4db6ac,stroke-width:2px,color:#fff;
    classDef detector fill:#b71c1c,stroke:#ef5350,stroke-width:2px,color:#fff;

    %% TERMINALS %%
    START([START])
    COMPLETE([COMPLETE])

    subgraph Step0 ["Step 0 — Setup"]
        direction TB
        ParseArgs["Parse Args<br/>━━━━━━━━━━<br/>research_question from<br/>skill invocation args"]
        DetectIssue{"GitHub issue ref<br/>━━━━━━━━━━<br/>URL / owner/repo#N / #N?"}
        FetchIssue["fetch_github_issue<br/>━━━━━━━━━━<br/>include_comments: true<br/>issue body → primary question"]
        MkDir["mkdir -p<br/>━━━━━━━━━━<br/>.autoskillit/temp/scope/"]
    end

    subgraph Step1 ["● Step 1 — Parallel Exploration (Task tool, model: sonnet)"]
        direction LR
        SubA["Subagent A<br/>━━━━━━━━━━<br/>Prior Art Survey<br/>existing impls + tests"]
        SubB["Subagent B<br/>━━━━━━━━━━<br/>Technical Context<br/>modules, data structures"]
        SubC["Subagent C<br/>━━━━━━━━━━<br/>External Research<br/>web search, tools, papers"]
        SubD["Subagent D<br/>━━━━━━━━━━<br/>Metric Context<br/>src/metrics.rs, thresholds"]
        SubE["★ Subagent E<br/>━━━━━━━━━━<br/>Computational Complexity<br/>O(n²)/O(n log n) analysis<br/>bottlenecks + gotchas"]
    end

    subgraph Step2 ["● Step 2 — Synthesize Findings"]
        direction TB
        Consolidate["Consolidate Subagent Outputs<br/>━━━━━━━━━━<br/>structured scope report"]
        subgraph ReportSections ["Output Template Sections"]
            direction LR
            S1["Research Question"]
            S2["Known/Unknown Matrix"]
            S3["Prior Art in Codebase"]
            S4["External Research"]
            S5["Technical Context"]
            S6["● Computational Complexity<br/>━━━━━━━━━━<br/>Dominant op · Scaling<br/>Bottlenecks · Gotchas"]
            S7["Hypotheses"]
            S8["Proposed Directions"]
            S9["Success Criteria"]
            S10["Metric Context"]
        end
    end

    subgraph Step3 ["Step 3 — Write Output"]
        direction TB
        SaveReport["Save Scope Report<br/>━━━━━━━━━━<br/>.autoskillit/temp/scope/<br/>scope_{topic}_{timestamp}.md"]
        EmitToken["Emit Output Token<br/>━━━━━━━━━━<br/>scope_report = {abs_path}"]
    end

    subgraph ContractTests ["★ tests/contracts/test_scope_contracts.py"]
        direction TB
        T1["test_section_exists<br/>━━━━━━━━━━<br/>## Computational Complexity<br/>present in SKILL.md"]
        T2["test_field_present<br/>━━━━━━━━━━<br/>Dominant op / Scaling /<br/>Known bottlenecks / Gotchas<br/>(parametrized ×4)"]
        T3["test_section_between<br/>━━━━━━━━━━<br/>TC_idx < CC_idx < Hyp_idx<br/>ordering enforced"]
        T4["test_baseline_computation<br/>━━━━━━━━━━<br/>regex: baseline.*computation<br/>in Known bottlenecks"]
        T5["test_complexity_subagent<br/>━━━━━━━━━━<br/>regex: Subagent.*Complexity<br/>in Step 1 instructions"]
    end

    %% FLOW %%
    START --> ParseArgs
    ParseArgs --> DetectIssue
    DetectIssue -->|"yes — issue ref found"| FetchIssue
    DetectIssue -->|"no — plain text question"| MkDir
    FetchIssue -->|"issue body as primary question"| MkDir

    MkDir --> SubA & SubB & SubC & SubD & SubE

    SubA & SubB & SubC & SubD & SubE --> Consolidate
    Consolidate --> S1 & S2 & S3 & S4 & S5 & S6 & S7 & S8 & S9 & S10
    S1 & S2 & S3 & S4 & S5 & S6 & S7 & S8 & S9 & S10 --> SaveReport

    SaveReport --> EmitToken
    EmitToken --> COMPLETE

    %% Contract tests read SKILL.md statically — not part of runtime flow
    T1 & T2 & T3 & T4 & T5 -.->|"reads SKILL.md at test-time"| S6

    %% CLASS ASSIGNMENTS %%
    class START,COMPLETE terminal;
    class ParseArgs,MkDir handler;
    class DetectIssue stateNode;
    class FetchIssue handler;
    class SubA,SubB,SubC,SubD phase;
    class SubE newComponent;
    class Consolidate phase;
    class S1,S2,S3,S4,S5,S7,S8,S9,S10 output;
    class S6 newComponent;
    class T1,T2,T3,T4,T5 detector;

Loading

Scenarios Diagram

%%{init: {'flowchart': {'nodeSpacing': 40, 'rankSpacing': 55, 'curve': 'basis'}}}%%
flowchart LR
    %% CLASS DEFINITIONS %%
    classDef cli fill:#1a237e,stroke:#7986cb,stroke-width:2px,color:#fff;
    classDef stateNode fill:#004d40,stroke:#4db6ac,stroke-width:2px,color:#fff;
    classDef handler fill:#e65100,stroke:#ffb74d,stroke-width:2px,color:#fff;
    classDef phase fill:#6a1b9a,stroke:#ba68c8,stroke-width:2px,color:#fff;
    classDef newComponent fill:#2e7d32,stroke:#81c784,stroke-width:2px,color:#fff;
    classDef output fill:#00695c,stroke:#4db6ac,stroke-width:2px,color:#fff;
    classDef detector fill:#b71c1c,stroke:#ef5350,stroke-width:2px,color:#fff;

    subgraph S1 ["SCENARIO 1: Scope Skill Happy Path"]
        direction LR
        S1_CLI["● scope/SKILL.md<br/>━━━━━━━━━━<br/>/autoskillit:scope {question}"]
        S1_ISSUE{"GitHub issue<br/>reference?"}
        S1_FETCH["fetch_github_issue<br/>━━━━━━━━━━<br/>include_comments: true"]
        S1_PARALLEL["Step 1: Parallel Subagents<br/>━━━━━━━━━━<br/>Task tool, model: sonnet"]
        S1_A["Subagent A<br/>━━━━━━━━━━<br/>Prior Art Survey"]
        S1_B["Subagent B<br/>━━━━━━━━━━<br/>Technical Context"]
        S1_C["Subagent C<br/>━━━━━━━━━━<br/>External Research"]
        S1_D["Subagent D<br/>━━━━━━━━━━<br/>Metric Context"]
        S1_E["★ Subagent E<br/>━━━━━━━━━━<br/>Computational Complexity<br/>(dominant op, scaling,<br/>bottlenecks, gotchas)"]
        S1_SYNTH["Step 2: Synthesize Findings<br/>━━━━━━━━━━<br/>Consolidate all subagents"]
        S1_OUT["scope report .md<br/>━━━━━━━━━━<br/>.autoskillit/temp/scope/"]
    end

    subgraph S2 ["SCENARIO 2: Contract Test Validation"]
        direction LR
        S2_PYTEST["pytest<br/>━━━━━━━━━━<br/>task test-check"]
        S2_TEST["★ test_scope_contracts.py<br/>━━━━━━━━━━<br/>TestComputationalComplexitySection"]
        S2_READ["_read_scope_skill_md()<br/>━━━━━━━━━━<br/>pkg_root() / skills_extended/scope/SKILL.md"]
        S2_CHECK1["test_section_exists<br/>━━━━━━━━━━<br/>## Computational Complexity in content"]
        S2_CHECK2["test_field_present<br/>━━━━━━━━━━<br/>Dominant op · Scaling ·<br/>Bottlenecks · Gotchas"]
        S2_CHECK3["test_section_between_...<br/>━━━━━━━━━━<br/>Tech Context &lt; CC &lt; Hypotheses"]
        S2_CHECK4["test_baseline_computation<br/>━━━━━━━━━━<br/>re.search baseline.*computation"]
        S2_CHECK5["test_complexity_subagent<br/>━━━━━━━━━━<br/>re.search Subagent.*Complexity"]
        S2_PASS["CI Green<br/>━━━━━━━━━━<br/>All 6 assertions pass"]
    end

    subgraph S3 ["SCENARIO 3: Report Output Structure"]
        direction LR
        S3_SYNTH["Synthesize step<br/>━━━━━━━━━━<br/>Consolidate subagent findings"]
        S3_TC["## Technical Context<br/>━━━━━━━━━━<br/>Architecture, modules, data flow"]
        S3_CC["● ## Computational Complexity<br/>━━━━━━━━━━<br/>Dominant operation<br/>Scaling behavior<br/>Known bottlenecks (+ baseline)<br/>Gotchas"]
        S3_HYP["## Hypotheses<br/>━━━━━━━━━━<br/>Falsifiable claims"]
        S3_FILE["scope_{topic}_{timestamp}.md<br/>━━━━━━━━━━<br/>Emits: scope_report = {abs_path}"]
    end

    %% SCENARIO 1 FLOW %%
    S1_CLI --> S1_ISSUE
    S1_ISSUE -- "yes" --> S1_FETCH
    S1_ISSUE -- "no" --> S1_PARALLEL
    S1_FETCH --> S1_PARALLEL
    S1_PARALLEL --> S1_A & S1_B & S1_C & S1_D & S1_E
    S1_A & S1_B & S1_C & S1_D & S1_E --> S1_SYNTH
    S1_SYNTH --> S1_OUT

    %% SCENARIO 2 FLOW %%
    S2_PYTEST --> S2_TEST
    S2_TEST --> S2_READ
    S2_READ --> S2_CHECK1 & S2_CHECK2 & S2_CHECK3 & S2_CHECK4 & S2_CHECK5
    S2_CHECK1 & S2_CHECK2 & S2_CHECK3 & S2_CHECK4 & S2_CHECK5 --> S2_PASS

    %% SCENARIO 3 FLOW %%
    S3_SYNTH --> S3_TC --> S3_CC --> S3_HYP --> S3_FILE

    %% CLASS ASSIGNMENTS %%
    class S1_CLI,S2_PYTEST stateNode;
    class S1_ISSUE detector;
    class S1_FETCH handler;
    class S1_PARALLEL,S1_SYNTH,S3_SYNTH phase;
    class S1_A,S1_B,S1_C,S1_D handler;
    class S1_E newComponent;
    class S1_OUT,S3_FILE output;
    class S2_TEST newComponent;
    class S2_READ,S2_CHECK1,S2_CHECK2,S2_CHECK3,S2_CHECK4,S2_CHECK5 handler;
    class S2_PASS output;
    class S3_TC handler;
    class S3_CC newComponent;
    class S3_HYP,S3_TC handler;

Loading

Closes #693

Implementation Plan

Plan file: /home/talon/projects/autoskillit-runs/impl-20260409-182034-729139/.autoskillit/temp/make-plan/scope_skill_computational_complexity_plan_2026-04-09_182600.md

🤖 Generated with Claude Code via AutoSkillit

Token Usage Summary

Step	uncached	output	cache_read	cache_write	count	time
plan	64	8.0k	855.8k	55.8k	1	5m 18s
verify	42	6.4k	533.0k	32.6k	1	4m 14s
implement	158	5.6k	667.9k	48.4k	1	2m 1s
prepare_pr	68	3.7k	211.6k	22.2k	1	1m 23s
run_arch_lenses	122	9.7k	392.9k	60.3k	2	3m 26s
compose_pr	67	6.6k	237.2k	27.8k	1	1m 48s
Total	521	40.0k	2.9M	247.0k		18m 13s

[Trecek]

[info] tests: _read_scope_skill_md() is called independently in every test method and parametrized case (7+ disk reads of the same immutable file). A class-level @pytest.fixture(scope='class') would eliminate redundant I/O with no functional change.

[Trecek]

AutoSkillit review found 2 blocking issues. See inline comments.