SC4SAP Analyze Code
Reviews an ABAP object by delegating the heavy work (source read, structural/semantic/where-used analysis, 14-dimension rule matching) to sap-code-reviewer (Opus 4.7). The main thread (Sonnet, per frontmatter) only handles Socratic intake, report formatting, and the follow-up action menu.
Invoke /sc4sap:trust-session with parent_skill=sc4sap:analyze-code to pre-grant all MCP tool + file-op permissions for this session (eliminates per-tool "Allow this tool?" prompts during the review flow).
- If
.sc4sap/session-trust.log already has a line within the last 24h, skip silently.
- Otherwise run it and surface the one-line confirmation.
Full spec: see ../trust-session/SKILL.md.
| Companion |
Scope |
analysis-dimensions.md |
9 common/ rule files that the reviewer agent (not main) loads + 14 evaluation dimensions |
workflow.md |
4-step execution flow: Identify → Review (delegated) → Report (branching) → Actions |
output-and-tools.md |
Report output format + MCP tool list used by the reviewer agent |
|
|
- Step 1 (main) — Identify: ask for (or confirm) the ABAP object name + type; verify via
SearchObject.
- Step 2 (delegated · Opus 4.7) — Dispatch to
sap-code-reviewer with only the object reference. The reviewer agent itself reads source (via GetClass/GetProgram/GetProgFullCode/...), runs structural analysis (GetAbapAST + GetAbapSemanticAnalysis + GetWhereUsed), loads the 9 common/ rule files, and evaluates all 14 dimensions. Returns: findings list (severity · location · rule ref · fix suggestion) + summary metrics.
- Step 3 (branching):
- Branch A — canned (default: no Critical findings AND < 10 findings total) → main formats the standard report template from
output-and-tools.md.
- Branch B — briefing (Critical present OR ≥ 10 findings) → main renders a richer reader-facing briefing (Critical/High with root cause + fix code, where-used impact, top-3 fixes) per
workflow.md § Branch B. No extra agent dispatch.
- Step 4 (main) — Follow-up action menu: show where-used · explain finding #N · save report (Markdown · HTML · both) · delegate fix to
sap-executor (user's choice).
Full spec in workflow.md. Main thread NEVER calls ReadClass / GetAbapAST / GetWhereUsed directly — that context stays inside the reviewer agent so the orchestrator window remains small even for large objects.
Task: {{ARGUMENTS}}
1---2name: sc4sap-analyze-code3description: ABAP code analysis — delegate source reads + AST/semantic/where-used analysis + rule-based review to sap-code-reviewer, then render a canned report or a richer briefing on the main thread4---56# SC4SAP Analyze Code78Reviews an ABAP object by delegating the heavy work (source read, structural/semantic/where-used analysis, 14-dimension rule matching) to `sap-code-reviewer` (Opus 4.7). The main thread (Sonnet, per frontmatter) only handles Socratic intake, report formatting, and the follow-up action menu.91011<Purpose>12sc4sap:analyze-code provides a comprehensive, severity-rated ABAP code review backed by the AST, semantic analysis, and where-used data that only the live SAP system can produce. The flow is deliberately thin on the main thread: the reviewer agent owns context-heavy work so the skill orchestrator stays light.13</Purpose>1415<Response_Prefix>16Every response triggered by this skill MUST begin with `[Model: <main-model> · Dispatched: <sub-summary>]` per [`../../common/model-routing-rule.md`](../../common/model-routing-rule.md) § Response Prefix Convention.17</Response_Prefix>1819<Phase_Banner>20Multi-phase skill. Before each `Agent(...)` dispatch, emit `▶ phase=<id> (<label>) · agent=<name> · model=<Opus 4.7|Sonnet 4.6|Haiku 4.5>` per [`../../common/model-routing-rule.md`](../../common/model-routing-rule.md) § Phase Banner Convention.21</Phase_Banner>2223<Team_Mode>24Type B (Coder ↔ Consultant) teamMode activates between Step 2 (reviewer full review) and Step 3 (report) when reviewer's findings include a **business-alignment dimension** (§ 1 Business Purpose, § 2 Rule Faithfulness, § 13 Cross-Module Side-Effects) AND the object touches 2+ modules. Reviewer becomes Worker, module consultants become Peers; consultants live-validate the reviewer's business-alignment findings. See [`team-mode.md`](team-mode.md). Base protocol: [`../../common/team-consultation-protocol.md`](../../common/team-consultation-protocol.md) § Type B.25</Team_Mode>2627<Use_When>28- User says "analyze", "review code", "check this class", "what's wrong with", "analyze code", or "code review"29- Before releasing a transport, to catch issues early30- When taking over existing ABAP code and wanting to understand its quality31- When optimizing performance of an ABAP program or class32- User wants where-used impact analysis before modifying an object33</Use_When>3435<Do_Not_Use_When>36- User wants to modify the code immediately → `/sc4sap:create-program` (full program flows) or direct `UpdateClass` / `UpdateProgram` / `UpdateInclude` MCP calls37- Object doesn't exist yet → `/sc4sap:create-object`38- User just wants to read the source → `ReadClass`, `ReadProgram` etc. directly39</Do_Not_Use_When>4041<Session_Trust_Bootstrap>42**MANDATORY — runs as Step 0 before any MCP call or user interaction.**4344Invoke `/sc4sap:trust-session` with `parent_skill=sc4sap:analyze-code` to pre-grant all MCP tool + file-op permissions for this session (eliminates per-tool "Allow this tool?" prompts during the review flow).4546- If `.sc4sap/session-trust.log` already has a line within the last 24h, skip silently.47- Otherwise run it and surface the one-line confirmation.4849Full spec: see [`../trust-session/SKILL.md`](../trust-session/SKILL.md).50</Session_Trust_Bootstrap>5152<Companion_Files>53**MANDATORY**: Read the companion files below before executing.5455| Companion | Scope |56|-----------|-------|57| [`analysis-dimensions.md`](analysis-dimensions.md) | 9 `common/` rule files that the **reviewer agent** (not main) loads + 14 evaluation dimensions |58| [`workflow.md`](workflow.md) | 4-step execution flow: Identify → Review (delegated) → Report (branching) → Actions |59| [`output-and-tools.md`](output-and-tools.md) | Report output format + MCP tool list used **by the reviewer agent** |60</Companion_Files>6162<Execution_Summary>63Orchestration is **1 main-thread Socratic intake + one delegated dispatch to `sap-code-reviewer` + a branching report + main-thread action menu**.6465- **Step 1 (main)** — Identify: ask for (or confirm) the ABAP object name + type; verify via `SearchObject`.66- **Step 2 (delegated · Opus 4.7)** — Dispatch to `sap-code-reviewer` with only the object reference. The reviewer agent **itself** reads source (via `GetClass`/`GetProgram`/`GetProgFullCode`/...), runs structural analysis (`GetAbapAST` + `GetAbapSemanticAnalysis` + `GetWhereUsed`), loads the 9 `common/` rule files, and evaluates all 14 dimensions. Returns: findings list (severity · location · rule ref · fix suggestion) + summary metrics.67- **Step 3 (branching)**:68 - **Branch A — canned** (default: no Critical findings AND < 10 findings total) → main formats the standard report template from [`output-and-tools.md`](output-and-tools.md).69 - **Branch B — briefing** (Critical present OR ≥ 10 findings) → main renders a richer reader-facing briefing (Critical/High with root cause + fix code, where-used impact, top-3 fixes) per [`workflow.md`](workflow.md) § Branch B. No extra agent dispatch.70- **Step 4 (main)** — Follow-up action menu: show where-used · explain finding #N · save report (Markdown · HTML · both) · delegate fix to `sap-executor` (user's choice).7172Full spec in [`workflow.md`](workflow.md). Main thread NEVER calls `ReadClass` / `GetAbapAST` / `GetWhereUsed` directly — that context stays inside the reviewer agent so the orchestrator window remains small even for large objects.73</Execution_Summary>7475Task: {{ARGUMENTS}}
babamba2