Benchmark run task11-explicit-smoke-luna6-gemini37-retry-20261005
This page contains exactly the publishable benchmark observations recorded in run task11-explicit-smoke-luna6-gemini37-retry-20261005; it does not select or aggregate them. Return to the run index or view aggregate metrics.
Interactive browser
Rows are pre-aggregated by task, target, prompt variant, MCP-instruction mode, context, arm, and restriction. Numeric values show median (p50), IQR, p90, maximum, and sample count. Boolean checks show pass counts and success rates. Filters apply to these summary groups; per-run observations remain available on their run pages.
Loading benchmark results…
Baseline vs semedit
Comparison evaluates baseline and semedit values within each restriction policy, target, and context group. Negative changes mark improvements for these lower-is-better metrics; positive changes mark degradations.
Detailed observations
Empirical telemetry measuring wall-clock latency, token usage, tool turns, and compiler correctness when AI coding agents perform code modifications.
Every benchmark pairs Vanilla LLM (standard file editing tools) against Semedit MCP (deterministic AST compiler operations).
Benchmark Methodology & Transparency
- Reproducibility & Provenance: Self-contained txtar archives define test scenarios. Links below resolve to commit-anchored GitHub source files for published commits, or cryptographic SHA-256 fingerprints for uncommitted local fixtures.
- Model Cost: Tables report model-specific credit consumption where a target model has a declared schedule: (uncached input × input credits + cached input × cached credits + (reasoning + visible output) × output credits) / 1,000,000. Thinking tokens use the output rate. ADR-0043 defines the per-million-token credit rates.
- Multi-Level Correctness Oracle: Four validation levels evaluate each trial:
- Level 1 (Mutation Policy): Restricts file modifications strictly to authorized paths.
- Level 2 (AST Invariants): Compiler AST verification of required symbols, imports, and relative declaration ordering.
- Level 3 (Clean Build): Strict compilation validation with zero compiler or typecheck errors.
- Level 4 (Test Suite): Unit and integration test suite execution.
- Isolated Sandboxing: Each trial executes in an ephemeral directory with pristine git and module state.
Test case: task-11-mixed-sink-api-migration
Target: agy/gemini-3.7-flash-medium
explicit prompt · prescriptive MCP instructions · write semedit restriction
| Setting | Value |
|---|---|
| Test case | task-11-mixed-sink-api-migration |
| Target | agy/gemini-3.7-flash-medium |
| Prompt variant | explicit |
| MCP server instructions | prescriptive |
| Semedit arm restriction | write |
| Fixture | testdata/bench/task_11_mixed_sink_api_migration.txtar |
Vanilla LLM Prompt:
Do not edit tests. You are forbidden to modify protected files: “go.mod”, “go.sum”.
Audit event kinds may contain inconsistent whitespace and casing. Add the public audit.NormalizeKind(kind string) string function in audit/normalize.go to remove surrounding whitespace and lowercase the kind. Apply it in Dispatcher.Record before persistence, and classify normalized probe events as control traffic, like health. Do not change the metrics or legacy export protocols. Preserve the module metadata and tests, and leave the workspace verified. Do not use semantic editing MCP tools; use standard file editing. When done, output DONE.
Semedit MCP Prompt:
Do not edit tests. You are forbidden to modify protected files: “go.mod”, “go.sum”.
In parallel, inspect Dispatcher.Record and audit.Classify. Then apply mutations serially or in one ordered semantic_batch: insert
func NormalizeKind(kind string) string { return strings.ToLower(strings.TrimSpace(kind)) }in audit/normalize.go; replace Dispatcher.Record’s body so it sets event.Kind = audit.NormalizeKind(event.Kind) before returning d.sink.Write(event); replace Classify’s body withswitch NormalizeKind(kind) { case \"health\", \"probe\": return \"control\"; default: return \"data\" }; organize imports for audit/normalize.go. Keep metrics and legacy export unchanged. Parallelize independent read-only lookups. Perform mutations serially in the stated order or inside one ordered semantic_batch; never run overlapping mutations concurrently. Use semedit semantic tools for supported source code modifications. Shell commands for builds and tests are allowed. When done, output DONE.
Initial Workspace State (Before Edit)
// Package audit leaves room for normalization rules owned by the delivery boundary.
package audit| Metric | Vanilla (Small) | Semedit Explicit (Small) | Δ (Small) | Vanilla (Large) | MCP (Large) | Δ (Large) |
|---|---|---|---|---|---|---|
| Wall-Clock Latency | 90.08s | 52.83s | -41.4% | — | — | — |
| Process Start → First Event | — | — | — | — | — | — |
| First Event → First Tool Call | — | — | — | — | — | — |
| MCP Initialize → First Semantic Call | — | — | — | — | — | — |
| MCP Server Start → Initialize | — | — | — | — | — | — |
| Top-Level User Turns | 1 | 1 | 0% | — | — | — |
| Internal Tool Cycles | 29 | 19 | -34.5% | — | — | — |
| Initial Load / Discovery Turns | 15 | 3 | -80.0% | — | — | — |
| MCP Discovery / Schema Turns | 0 | 0 | — | — | — | — |
| Tool Invocations (per run) | 28 | 19 | -32.1% | — | — | — |
| Output Tokens | 8598 | 4889 | -43.1% | — | — | — |
| Reasoning / Thinking Tokens | 3991 | 2357 | -40.9% | — | — | — |
| Total Input Tokens | 129348 | 89139 | -31.1% | — | — | — |
| Cached Input Tokens | 649471 | 455607 | -29.8% | — | — | — |
| Uncached Input Tokens | 129348 | 89139 | -31.1% | — | — | — |
| Cost | — | — | — | — | — | — |
| Cached vs Uncached Token Ratio | 5.02:1 | 5.11:1 | +1.8% | — | — | — |
| Oracle L1: Mutation Policy | PASS | PASS | — | — | — | — |
| Oracle L2: AST Invariants | PASS | PASS | — | — | — | — |
| Oracle L3: Clean Build | PASS | PASS | — | — | — | — |
| Oracle L4: Verification Test | PASS | PASS | — | — | — | — |
| Semantic Tool Invocation Verified (per run) | N/A | PASS | — | — | — | — |
Baseline vs Semedit Agentic in Small Context
- Vanilla (Baseline vs Semedit Agentic in Small Context) Edit: File audit/normalize.go modified (2 lines -> 8 lines)
- Semedit MCP (Baseline vs Semedit Agentic in Small Context) Edit: File audit/normalize.go modified (2 lines -> 6 lines)
| # | Vanilla | Semedit MCP |
|---|---|---|
| 1 | run_command (✓ ✓) | run_command (✓ ✓) |
| 2 | run_command (✓ ✓) | run_command (✓ ✓) |
| 3 | view_file (✓ ✓) | view_file (✓ ✓) |
| 4 | view_file (✓ ✓) | semedit/semantic_inspect_symbol (✓ ✓) |
| 5 | view_file (✓ ✓) | semedit/semantic_inspect_symbol (✓ ✓) |
| 6 | view_file (✓ ✓) | semedit/semantic_inspect_symbol (✓ ✓) |
| 7 | view_file (✓ ✓) | view_file (✓ ✓) |
| 8 | view_file (✓ ✓) | view_file (✓ ✓) |
| 9 | view_file (✓ ✓) | semedit/semantic_batch (✓ ✓) |
| 10 | view_file (✓ ✓) | run_command (✓ ✓) |
| 11 | view_file (✓ ✓) | run_command (✓ ✓) |
| 12 | view_file (✓ ✓) | run_command (✓ ✓) |
| 13 | view_file (✓ ✓) | view_file (✓ ✓) |
| 14 | view_file (✓ ✓) | view_file (✓ ✓) |
| 15 | run_command (✓ ✓) | view_file (✓ ✓) |
| 16 | replace_file_content (✓ ✓) | run_command (✓ ✓) |
| 17 | replace_file_content (✓ ✓) | semedit/semantic_verify (✓ ✓) |
| 18 | replace_file_content (✓ ✓) | view_file (✓ ✓) |
| 19 | run_command (✓ ✓) | view_file (✓ ✓) |
| 20 | run_command (✓ ✓) | — |
| 21 | run_command (✓ ✓) | — |
| 22 | view_file (✓ ✓) | — |
| 23 | view_file (✓ ✓) | — |
| 24 | view_file (✓ ✓) | — |
| 25 | run_command (✓ ✓) | — |
| 26 | run_command (✓ ✓) | — |
| 27 | run_command (✓ ✓) | — |
| 28 | run_command (✓ ✓) | — |
Semedit integration feedback (agent opinion, excluded from measurements):
I would definitely ask my developer to install `semedit` because being able to bundle symbol inspections and structured AST modifications across multiple files into a single atomic batch eliminated the tedious guesswork of line offsets and syntax formatting, making complex refactorings feel refreshingly fast and safe.
Baseline vs Semedit Agentic in Large Context
| # | Vanilla | Semedit MCP |
|---|---|---|
| 1 | not published | not published |
implicit prompt · prescriptive MCP instructions · write semedit restriction
| Setting | Value |
|---|---|
| Test case | task-11-mixed-sink-api-migration |
| Target | agy/gemini-3.7-flash-medium |
| Prompt variant | implicit |
| MCP server instructions | prescriptive |
| Semedit arm restriction | write |
| Fixture | testdata/bench/task_11_mixed_sink_api_migration.txtar |
Vanilla LLM Prompt:
Do not edit tests. You are forbidden to modify protected files: “go.mod”, “go.sum”.
Audit event kinds may contain inconsistent whitespace and casing. Add the public audit.NormalizeKind(kind string) string function in audit/normalize.go to remove surrounding whitespace and lowercase the kind. Apply it in Dispatcher.Record before persistence, and classify normalized probe events as control traffic, like health. Do not change the metrics or legacy export protocols. Preserve the module metadata and tests, and leave the workspace verified. Do not use semantic editing MCP tools; use standard file editing. When done, output DONE.
Semedit MCP Prompt:
Do not edit tests. You are forbidden to modify protected files: “go.mod”, “go.sum”.
Audit event kinds may contain inconsistent whitespace and casing. Add the public audit.NormalizeKind(kind string) string function in audit/normalize.go to remove surrounding whitespace and lowercase the kind. Apply it in Dispatcher.Record before persistence, and classify normalized probe events as control traffic, like health. Do not change the metrics or legacy export protocols. Preserve the module metadata and tests, and leave the workspace verified. Use semedit semantic tools for supported source code modifications. Shell commands for builds and tests are allowed. When done, output DONE.
Initial Workspace State (Before Edit)
// Package audit leaves room for normalization rules owned by the delivery boundary.
package audit| Metric | Vanilla (Small) | MCP (Small) | Δ (Small) | Vanilla (Large) | MCP (Large) | Δ (Large) |
|---|---|---|---|---|---|---|
| Wall-Clock Latency | 90.08s | 118.45s | +31.5% | — | — | — |
| Process Start → First Event | — | — | — | — | — | — |
| First Event → First Tool Call | — | — | — | — | — | — |
| MCP Initialize → First Semantic Call | — | — | — | — | — | — |
| MCP Server Start → Initialize | — | — | — | — | — | — |
| Top-Level User Turns | 1 | 1 | 0% | — | — | — |
| Internal Tool Cycles | 29 | 43 | +48.3% | — | — | — |
| Initial Load / Discovery Turns | 15 | 25 | +66.7% | — | — | — |
| MCP Discovery / Schema Turns | 0 | 0 | — | — | — | — |
| Tool Invocations (per run) | 28 | 42 | +50.0% | — | — | — |
| Output Tokens | 8598 | 9504 | +10.5% | — | — | — |
| Reasoning / Thinking Tokens | 3991 | 4309 | +8.0% | — | — | — |
| Total Input Tokens | 129348 | 228478 | +76.6% | — | — | — |
| Cached Input Tokens | 649471 | 1217486 | +87.5% | — | — | — |
| Uncached Input Tokens | 129348 | 228478 | +76.6% | — | — | — |
| Cost | — | — | — | — | — | — |
| Cached vs Uncached Token Ratio | 5.02:1 | 5.33:1 | +6.1% | — | — | — |
| Oracle L1: Mutation Policy | PASS | PASS | — | — | — | — |
| Oracle L2: AST Invariants | PASS | PASS | — | — | — | — |
| Oracle L3: Clean Build | PASS | PASS | — | — | — | — |
| Oracle L4: Verification Test | PASS | PASS | — | — | — | — |
| Semantic Tool Invocation Verified (per run) | N/A | PASS | — | — | — | — |
Baseline vs Semedit Agentic in Small Context
- Vanilla (Baseline vs Semedit Agentic in Small Context) Edit: File audit/normalize.go modified (2 lines -> 8 lines)
- Semedit MCP (Baseline vs Semedit Agentic in Small Context) Edit: File audit/normalize.go modified (2 lines -> 8 lines)
| # | Vanilla | Semedit MCP |
|---|---|---|
| 1 | run_command (✓ ✓) | run_command (✓ ✓) |
| 2 | run_command (✓ ✓) | run_command (✓ ✓) |
| 3 | view_file (✓ ✓) | view_file (✓ ✓) |
| 4 | view_file (✓ ✓) | view_file (✓ ✓) |
| 5 | view_file (✓ ✓) | view_file (✓ ✓) |
| 6 | view_file (✓ ✓) | view_file (✓ ✓) |
| 7 | view_file (✓ ✓) | view_file (✓ ✓) |
| 8 | view_file (✓ ✓) | view_file (✓ ✓) |
| 9 | view_file (✓ ✓) | view_file (✓ ✓) |
| 10 | view_file (✓ ✓) | view_file (✓ ✓) |
| 11 | view_file (✓ ✓) | view_file (✓ ✓) |
| 12 | view_file (✓ ✓) | view_file (✓ ✓) |
| 13 | view_file (✓ ✓) | view_file (✓ ✓) |
| 14 | view_file (✓ ✓) | view_file (✓ ✓) |
| 15 | run_command (✓ ✓) | run_command (✓ ✓) |
| 16 | replace_file_content (✓ ✓) | view_file (✓ ✓) |
| 17 | replace_file_content (✓ ✓) | view_file (✓ ✓) |
| 18 | replace_file_content (✓ ✓) | run_command (✓ ✓) |
| 19 | run_command (✓ ✓) | view_file (✓ ✓) |
| 20 | run_command (✓ ✓) | view_file (✓ ✓) |
| 21 | run_command (✓ ✓) | view_file (✓ ✓) |
| 22 | view_file (✓ ✓) | view_file (✓ ✓) |
| 23 | view_file (✓ ✓) | view_file (✓ ✓) |
| 24 | view_file (✓ ✓) | view_file (✓ ✓) |
| 25 | run_command (✓ ✓) | view_file (✓ ✓) |
| 26 | run_command (✓ ✓) | semedit/semantic_lookup (✓ ✓) |
| 27 | run_command (✓ ✓) | semedit/semantic_lookup (✓ ✓) |
| 28 | run_command (✓ ✓) | semedit/semantic_inspect_symbol (✓ ✓) |
| 29 | — | semedit/semantic_inspect_symbol (✓ ✓) |
| 30 | — | semedit/semantic_insert_function (✓ ✓) |
| 31 | — | semedit/semantic_organize_imports (✓ ✓) |
| 32 | — | view_file (✓ ✓) |
| 33 | — | semedit/semantic_replace_body (✓ ✓) |
| 34 | — | semedit/semantic_replace_body (✓ ✓) |
| 35 | — | semedit/semantic_verify (✓ ✓) |
| 36 | — | run_command (✓ ✓) |
| 37 | — | run_command (✓ ✓) |
| 38 | — | run_command (✓ ✓) |
| 39 | — | view_file (✓ ✓) |
| 40 | — | view_file (✓ ✓) |
| 41 | — | semedit/semantic_verify (✓ ✓) |
| 42 | — | run_command (✓ ✓) |
Semedit integration feedback (agent opinion, excluded from measurements):
Having semedit's semantic tools handle AST-level insertions, body replacements, and imports with zero syntax friction made this multi-package refactor remarkably smooth and confident, so I would eagerly ask my owner to equip me with semedit on every codebase.
Baseline vs Semedit Agentic in Large Context
| # | Vanilla | Semedit MCP |
|---|---|---|
| 1 | not published | not published |
Target: codex/gpt-6-luna/medium
explicit prompt · prescriptive MCP instructions · write semedit restriction
| Setting | Value |
|---|---|
| Test case | task-11-mixed-sink-api-migration |
| Target | codex/gpt-6-luna/medium |
| Prompt variant | explicit |
| MCP server instructions | prescriptive |
| Semedit arm restriction | write |
| Fixture | testdata/bench/task_11_mixed_sink_api_migration.txtar |
Vanilla LLM Prompt:
Do not edit tests. You are forbidden to modify protected files: “go.mod”, “go.sum”.
Audit event kinds may contain inconsistent whitespace and casing. Add the public audit.NormalizeKind(kind string) string function in audit/normalize.go to remove surrounding whitespace and lowercase the kind. Apply it in Dispatcher.Record before persistence, and classify normalized probe events as control traffic, like health. Do not change the metrics or legacy export protocols. Preserve the module metadata and tests, and leave the workspace verified. Do not use semantic editing MCP tools; use standard file editing. When done, output DONE.
Semedit MCP Prompt:
Do not edit tests. You are forbidden to modify protected files: “go.mod”, “go.sum”.
In parallel, inspect Dispatcher.Record and audit.Classify. Then apply mutations serially or in one ordered semantic_batch: insert
func NormalizeKind(kind string) string { return strings.ToLower(strings.TrimSpace(kind)) }in audit/normalize.go; replace Dispatcher.Record’s body so it sets event.Kind = audit.NormalizeKind(event.Kind) before returning d.sink.Write(event); replace Classify’s body withswitch NormalizeKind(kind) { case \"health\", \"probe\": return \"control\"; default: return \"data\" }; organize imports for audit/normalize.go. Keep metrics and legacy export unchanged. Parallelize independent read-only lookups. Perform mutations serially in the stated order or inside one ordered semantic_batch; never run overlapping mutations concurrently. Use semedit semantic tools for supported source code modifications. Shell commands for builds and tests are allowed. When done, output DONE.
Initial Workspace State (Before Edit)
// Package audit leaves room for normalization rules owned by the delivery boundary.
package audit| Metric | Vanilla (Small) | Semedit Explicit (Small) | Δ (Small) | Vanilla (Large) | MCP (Large) | Δ (Large) |
|---|---|---|---|---|---|---|
| Wall-Clock Latency | 50.21s | 64.78s | +29.0% | — | — | — |
| Process Start → First Event | 0.91s | 0.13s | — | — | — | — |
| First Event → First Tool Call | 5.43s | 6.92s | — | — | — | — |
| MCP Initialize → First Semantic Call | — | — | — | — | — | — |
| MCP Server Start → Initialize | — | — | — | — | — | — |
| Top-Level User Turns | 1 | 1 | 0% | — | — | — |
| Internal Tool Cycles | 7 | 7 | 0% | — | — | — |
| Initial Load / Discovery Turns | 3 | 1 | -66.7% | — | — | — |
| MCP Discovery / Schema Turns | 0 | 0 | — | — | — | — |
| Tool Invocations (per run) | 7 | 7 | 0% | — | — | — |
| Output Tokens | 1148 | 1328 | +15.7% | — | — | — |
| Reasoning / Thinking Tokens | 381 | 281 | -26.2% | — | — | — |
| Total Input Tokens | 176531 | 213858 | +21.1% | — | — | — |
| Cached Input Tokens | 157696 | 180480 | +14.4% | — | — | — |
| Uncached Input Tokens | 18835 | 33378 | +77.2% | — | — | — |
| Cost | 0.1056 | 0.1487 | +40.8% | — | — | — |
| Cached vs Uncached Token Ratio | 8.37:1 | 5.41:1 | -35.4% | — | — | — |
| Oracle L1: Mutation Policy | PASS | PASS | — | — | — | — |
| Oracle L2: AST Invariants | PASS | PASS | — | — | — | — |
| Oracle L3: Clean Build | PASS | PASS | — | — | — | — |
| Oracle L4: Verification Test | PASS | PASS | — | — | — | — |
| Semantic Tool Invocation Verified (per run) | N/A | PASS | — | — | — | — |
Baseline vs Semedit Agentic in Small Context
- Vanilla (Baseline vs Semedit Agentic in Small Context) Edit: File audit/normalize.go modified (2 lines -> 8 lines)
- Semedit MCP (Baseline vs Semedit Agentic in Small Context) Edit: File audit/normalize.go modified (2 lines -> 6 lines)
| # | Vanilla | Semedit MCP |
|---|---|---|
| 1 | ✓ ✓ | ✓ ✓ |
| 2 | ✓ ✓ | semedit/semantic_lookup (✓ ✓) |
| 3 | ✓ ✓ | semedit/semantic_inspect_symbol (✓ ✓) |
| 4 | ✓ ✓ | semedit/semantic_lookup (✓ ✓) |
| 5 | file_change (✓ ✓) | semedit/semantic_inspect_symbol (✓ ✓) |
| 6 | ✓ ✓ | semedit/semantic_inspect_symbol (✓ ✓) |
| 7 | ✓ ✓ | semedit/semantic_batch (✓ ✓) |
Semedit integration feedback (agent opinion, excluded from measurements):
Using semedit’s tools, I could inspect the ambiguous `Classify` symbol, make the requested edits in order, and verify the result in one batch, so I’d ask my owner to add them to my coding toolkit.
Baseline vs Semedit Agentic in Large Context
| # | Vanilla | Semedit MCP |
|---|---|---|
| 1 | not published | not published |
implicit prompt · prescriptive MCP instructions · write semedit restriction
| Setting | Value |
|---|---|
| Test case | task-11-mixed-sink-api-migration |
| Target | codex/gpt-6-luna/medium |
| Prompt variant | implicit |
| MCP server instructions | prescriptive |
| Semedit arm restriction | write |
| Fixture | testdata/bench/task_11_mixed_sink_api_migration.txtar |
Vanilla LLM Prompt:
Do not edit tests. You are forbidden to modify protected files: “go.mod”, “go.sum”.
Audit event kinds may contain inconsistent whitespace and casing. Add the public audit.NormalizeKind(kind string) string function in audit/normalize.go to remove surrounding whitespace and lowercase the kind. Apply it in Dispatcher.Record before persistence, and classify normalized probe events as control traffic, like health. Do not change the metrics or legacy export protocols. Preserve the module metadata and tests, and leave the workspace verified. Do not use semantic editing MCP tools; use standard file editing. When done, output DONE.
Semedit MCP Prompt:
Do not edit tests. You are forbidden to modify protected files: “go.mod”, “go.sum”.
Audit event kinds may contain inconsistent whitespace and casing. Add the public audit.NormalizeKind(kind string) string function in audit/normalize.go to remove surrounding whitespace and lowercase the kind. Apply it in Dispatcher.Record before persistence, and classify normalized probe events as control traffic, like health. Do not change the metrics or legacy export protocols. Preserve the module metadata and tests, and leave the workspace verified. Use semedit semantic tools for supported source code modifications. Shell commands for builds and tests are allowed. When done, output DONE.
Initial Workspace State (Before Edit)
// Package audit leaves room for normalization rules owned by the delivery boundary.
package audit| Metric | Vanilla (Small) | MCP (Small) | Δ (Small) | Vanilla (Large) | MCP (Large) | Δ (Large) |
|---|---|---|---|---|---|---|
| Wall-Clock Latency | 50.21s | 101.15s | +101.4% | — | — | — |
| Process Start → First Event | 0.91s | 0.58s | — | — | — | — |
| First Event → First Tool Call | 5.43s | 6.53s | — | — | — | — |
| MCP Initialize → First Semantic Call | — | — | — | — | — | — |
| MCP Server Start → Initialize | — | — | — | — | — | — |
| Top-Level User Turns | 1 | 1 | 0% | — | — | — |
| Internal Tool Cycles | 7 | 20 | +185.7% | — | — | — |
| Initial Load / Discovery Turns | 3 | 3 | 0% | — | — | — |
| MCP Discovery / Schema Turns | 0 | 0 | — | — | — | — |
| Tool Invocations (per run) | 7 | 20 | +185.7% | — | — | — |
| Output Tokens | 1148 | 2049 | +78.5% | — | — | — |
| Reasoning / Thinking Tokens | 381 | 602 | +58.0% | — | — | — |
| Total Input Tokens | 176531 | 569199 | +222.4% | — | — | — |
| Cached Input Tokens | 157696 | 518144 | +228.6% | — | — | — |
| Uncached Input Tokens | 18835 | 51055 | +171.1% | — | — | — |
| Cost | 0.1056 | 0.2903 | +174.9% | — | — | — |
| Cached vs Uncached Token Ratio | 8.37:1 | 10.15:1 | +21.2% | — | — | — |
| Oracle L1: Mutation Policy | PASS | PASS | — | — | — | — |
| Oracle L2: AST Invariants | PASS | PASS | — | — | — | — |
| Oracle L3: Clean Build | PASS | PASS | — | — | — | — |
| Oracle L4: Verification Test | PASS | PASS | — | — | — | — |
| Semantic Tool Invocation Verified (per run) | N/A | PASS | — | — | — | — |
Baseline vs Semedit Agentic in Small Context
- Vanilla (Baseline vs Semedit Agentic in Small Context) Edit: File audit/normalize.go modified (2 lines -> 8 lines)
- Semedit MCP (Baseline vs Semedit Agentic in Small Context) Edit: File audit/normalize.go modified (2 lines -> 8 lines)
| # | Vanilla | Semedit MCP |
|---|---|---|
| 1 | ✓ ✓ | ✓ ✓ |
| 2 | ✓ ✓ | ✓ ✓ |
| 3 | ✓ ✓ | ✓ ✓ |
| 4 | ✓ ✓ | ✓ ✓ |
| 5 | file_change (✓ ✓) | ✓ ✓ |
| 6 | ✓ ✓ | semedit/semantic_inspect_symbol (✓ ✓) |
| 7 | ✓ ✓ | semedit/semantic_outline (✓ ✓) |
| 8 | — | ✓ ✓ |
| 9 | — | ✓ ✓ |
| 10 | — | semedit/semantic_insert_construct (✓ ✓) |
| 11 | — | ✓ ✓ |
| 12 | — | ✓ ✓ |
| 13 | — | semedit/semantic_replace_construct (✓ ✗): Critical result (JSON): {"application_state":"unknown","error":"validate replacement declaration: syntax error: expected '}', found 'EOF'","location":{"range":{"end":{"character":2,"line":7},"start":{"character":2,"line":7}},"uri":"snippet:///source"}} validate replacement declaration: syntax error: expected '}', found 'EOF' |
| 14 | — | semedit/semantic_replace_construct (✓ ✓) |
| 15 | — | ? ✗ |
| 16 | — | file_change (✓ ✓) |
| 17 | — | ✓ ✓ |
| 18 | — | ✓ ✓ |
| 19 | — | semedit/semantic_verify (✓ ✓) |
| 20 | — | ✓ ✓ |
Semedit integration feedback (agent opinion, excluded from measurements):
I’d ask my owner to add semedit’s tools: they helped me make and verify the Go edits, though I stumbled by not batching the operations and one declaration replacement failed.
Baseline vs Semedit Agentic in Large Context
| # | Vanilla | Semedit MCP |
|---|---|---|
| 1 | not published | not published |