Add intrinsic prompt-pack intervention for raw generation (sprint 261)
This commit is contained in:
Bill
@@ -237,3 +237,8 @@ Planning runtime controls (active):
|
||||
- `WSTONE_NATIVE_RAW_SIGNAL_TARGETED_VARIANTS`
|
||||
- `WSTONE_NATIVE_RAW_SIGNAL_TARGETED_MAX_VARIANTS`
|
||||
- `native_raw_signal_targeted_variants`
|
||||
- raw candidate generation can inject intrinsic prompt-pack constraints from gap signals:
|
||||
- `tools/mcp/synthesize_raw_intrinsic_prompt_pack.py`
|
||||
- `WSTONE_NATIVE_RAW_INTRINSIC_PROMPT_PACK`
|
||||
- `WSTONE_NATIVE_RAW_INTRINSIC_PROMPT_MAX_SIGNALS`
|
||||
- `native_raw_intrinsic_prompt_pack`
|
||||
|
||||
@@ -47,7 +47,7 @@ This is the canonical dated registry for "not production-ready" generator gaps.
|
||||
| GR-018 | Performance/security/rollout constrained refactor enforcement gap | `docs/gap_hunt_fullstack_multifile_2026-02-26.md`, `logs/taskitem_runs/challenging_fullstack_multifile_20260226_r5/results.jsonl` | `partial` | Hard checks now enforce migration rollback + data-loss policy, security deny-by-default, SLO p95 presence, and rollout staged+abort policy. Enforcement is contract-level; generator capability under these constraints is still weak in C++ AB path. | Constraint-aware generation follow-up |
|
||||
| GR-019 | Parity-blocked readiness load (gating without capability closure) | `logs/taskitem_runs/challenging_fullstack_multifile_20260226_r7/summary.json`, `logs/taskitem_runs/challenging_subset_prod_20260226_r7/summary.json`, `docs/sprint225_227_execution_tracker_2026-02-26.md` | `partial` | Sprint 225-227 reduced blocked parity load from `6` to `0` on both tracked hard catalogs while keeping unresolved divergence at `0`. This closes immediate safety debt for current corpora, but robustness is still contingent on pattern-driven repair classes. | Generalize repairs beyond queue-shaped transpile outputs |
|
||||
| GR-020 | Semantic fallback overuse masks weak native decomposition | `logs/taskitem_runs/TEST_ONLY_sprint236_semantic_fallback_audit_20260226/semantic_fallback_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint237_semantic_fallback_gate_fail_20260226/semantic_fallback_budget_gate.json`, `logs/taskitem_runs/TEST_ONLY_sprint238_native_gate_fail_20260226.json`, `logs/taskitem_runs/TEST_ONLY_sprint239_semantic_fallback_audit_20260226/semantic_fallback_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint240_retry_20260226_145631/00_summary.json`, `docs/sprint236_execution_tracker_2026-02-26.md`, `docs/sprint237_execution_tracker_2026-02-26.md`, `docs/sprint238_execution_tracker_2026-02-26.md`, `docs/sprint239_execution_tracker_2026-02-26.md`, `docs/sprint240_execution_tracker_2026-02-26.md` | `partial` | Sprint 236 added deterministic fallback-gap auditing with tool + constraint metadata. Sprint 237 added fallback-budget hard gate (`max-fallback-rate`) and produced expected fail/pass artifacts. Sprint 238 added native decomposition hard gate in pipeline (`min task count`, `min semantic signal count`) so weak native generation can be blocked before fallback masking. Sprint 239 added native reason enrichment and improved semantic signal density (`0 -> 6`). Sprint 240 added native decomposition retry with explicit minimum-task policy; on current sample retry attempted but did not improve depth (`2 -> 2`). | Native decomposition task-depth upgrade (increase native task granularity beyond 2) |
|
||||
| GR-021 | Impact-specific native decomposition coverage not enforced uniformly | `docs/native_decomposition_impact_list_2026-02-26.md`, `tools/mcp/profiles/native_decomposition_impact_profiles.json`, `logs/taskitem_runs/TEST_ONLY_sprint241_fullstack_impact_20260226_150416/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint241_native_impact_aggregate_20260226/native_impact_coverage_aggregate.json`, `logs/taskitem_runs/TEST_ONLY_sprint242_impact_remediation_loop_20260226/remediation_loop_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint243_impact_remediation_tasks_loop_20260226_r2/remediation_loop_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint244_autofill_20260226_151651/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint244_autofill_enforce_20260226_151709/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint245_intrinsic_20260226_151835/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint246_multishot_20260226_153033/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint246_multishot_enforce_20260226_153045/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint247_singleshot_20260226_153230/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint247_singleshot_enforce_20260226_153241/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint248_rawsearch_20260226_153903/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint249_rawguard_20260226_154028/02ae_raw_candidate_search.json`, `logs/taskitem_runs/TEST_ONLY_sprint250_closure_ladder_20260226/closure_ladder_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint251_rawvariants_20260226_154355/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint252_closure_ladder_policy_skip2_20260226/closure_ladder_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint253_closure_ladder_batch_20260226/closure_ladder_batch_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint254_closure_ladder_batch_20260226_r2/raw_gap_backlog.json`, `logs/taskitem_runs/TEST_ONLY_sprint255_rawharden_20260226_161924/00_summary.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_162423/00_summary.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_162633/02ae_raw_candidate_search.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_162924/01e_raw_top_gap_requirements_report.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163148/02af_raw_adaptive_retry_score.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163556/02ae_raw_signal_targeted_variants.json` | `partial` | Sprint 241 introduced impact profiles/gates; 242-243 remediation loops; 244 first-pass autofill; 245 intrinsic constraints only (no uplift); 246 multishot closure; 247 single-shot shaping closure; 248 raw candidate search no uplift; 249 no-uplift guardrail; 250 closure ladder; 251 richer raw variants no uplift; 252 history-aware ladder routing; 253 batch ladder analytics; 254 raw gap backlog synthesis from `raw_only` attempts identifies top missing raw signals; 255 adds first-pass raw top-gap hardening that closes hard-sample profile coverage (`failing_profile_count 7 -> 0`) by injecting missing ops/contracts/reasons and minimum task depth; 256 adds top-gap weighted candidate selection telemetry and tie-break policy; 257 adds hard fail policy when top-gap uplift is required but absent (`exit 18`); 258 injects backlog-driven top-gap normalized requirements before raw generation; 259 adds adaptive retry with expanded top-gap requirements and deterministic retry scoring; 260 adds signal-targeted variant expansion keyed to missing-signal classes and increases intrinsic search breadth (`available_variants 2 -> 8`). Measured intrinsic output still shows no uplift on this sample (`best_top_gap_score=0`, `failing_profile_count=1`). | Add model/prompt-level intrinsic generation intervention keyed by missing-signal classes (beyond requirement text injection alone) |
|
||||
| GR-021 | Impact-specific native decomposition coverage not enforced uniformly | `docs/native_decomposition_impact_list_2026-02-26.md`, `tools/mcp/profiles/native_decomposition_impact_profiles.json`, `logs/taskitem_runs/TEST_ONLY_sprint241_fullstack_impact_20260226_150416/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint241_native_impact_aggregate_20260226/native_impact_coverage_aggregate.json`, `logs/taskitem_runs/TEST_ONLY_sprint242_impact_remediation_loop_20260226/remediation_loop_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint243_impact_remediation_tasks_loop_20260226_r2/remediation_loop_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint244_autofill_20260226_151651/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint244_autofill_enforce_20260226_151709/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint245_intrinsic_20260226_151835/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint246_multishot_20260226_153033/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint246_multishot_enforce_20260226_153045/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint247_singleshot_20260226_153230/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint247_singleshot_enforce_20260226_153241/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint248_rawsearch_20260226_153903/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint249_rawguard_20260226_154028/02ae_raw_candidate_search.json`, `logs/taskitem_runs/TEST_ONLY_sprint250_closure_ladder_20260226/closure_ladder_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint251_rawvariants_20260226_154355/00_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint252_closure_ladder_policy_skip2_20260226/closure_ladder_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint253_closure_ladder_batch_20260226/closure_ladder_batch_summary.json`, `logs/taskitem_runs/TEST_ONLY_sprint254_closure_ladder_batch_20260226_r2/raw_gap_backlog.json`, `logs/taskitem_runs/TEST_ONLY_sprint255_rawharden_20260226_161924/00_summary.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_162423/00_summary.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_162633/02ae_raw_candidate_search.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_162924/01e_raw_top_gap_requirements_report.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163148/02af_raw_adaptive_retry_score.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163556/02ae_raw_signal_targeted_variants.json`, `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163855/01f_raw_intrinsic_prompt_pack_report.json` | `partial` | Sprint 241 introduced impact profiles/gates; 242-243 remediation loops; 244 first-pass autofill; 245 intrinsic constraints only (no uplift); 246 multishot closure; 247 single-shot shaping closure; 248 raw candidate search no uplift; 249 no-uplift guardrail; 250 closure ladder; 251 richer raw variants no uplift; 252 history-aware ladder routing; 253 batch ladder analytics; 254 raw gap backlog synthesis from `raw_only` attempts identifies top missing raw signals; 255 adds first-pass raw top-gap hardening that closes hard-sample profile coverage (`failing_profile_count 7 -> 0`) by injecting missing ops/contracts/reasons and minimum task depth; 256 adds top-gap weighted candidate selection telemetry and tie-break policy; 257 adds hard fail policy when top-gap uplift is required but absent (`exit 18`); 258 injects backlog-driven top-gap normalized requirements before raw generation; 259 adds adaptive retry with expanded top-gap requirements and deterministic retry scoring; 260 adds signal-targeted variant expansion keyed to missing-signal classes and increases intrinsic search breadth (`available_variants 2 -> 8`); 261 adds intrinsic prompt-pack constraint injection keyed to gap signals. Measured intrinsic output still shows no uplift on this sample (`best_top_gap_score=0`, `failing_profile_count=1`). | Add explicit generator-side output-shape control (schema-guided decoder or task-template synthesis pass) before candidate scoring |
|
||||
|
||||
## What Was Covered Today (Sprints 175-184)
|
||||
|
||||
|
||||
@@ -490,3 +490,20 @@
|
||||
- Current measured result on hard sample:
|
||||
- search breadth increased (`available_variants 2 -> 8`)
|
||||
- intrinsic outcome unchanged (`selected_variant=0`, `best_top_gap_score=0`, `failing_profile_count=1`).
|
||||
|
||||
## Sprint 261 Added (Same Day)
|
||||
|
||||
- Added intrinsic prompt-pack intervention:
|
||||
- `tools/mcp/synthesize_raw_intrinsic_prompt_pack.py`
|
||||
- pipeline controls:
|
||||
- `WSTONE_NATIVE_RAW_INTRINSIC_PROMPT_PACK`
|
||||
- `WSTONE_NATIVE_RAW_INTRINSIC_PROMPT_MAX_SIGNALS`
|
||||
- summary packet:
|
||||
- `native_raw_intrinsic_prompt_pack`
|
||||
- Dated A/B artifacts:
|
||||
- OFF: `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163844/00_summary.json`
|
||||
- ON: `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163855/00_summary.json`
|
||||
- report: `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163855/01f_raw_intrinsic_prompt_pack_report.json`
|
||||
- Current measured result on hard sample:
|
||||
- prompt pack injected (`selected_signal_count=8`)
|
||||
- intrinsic outcome unchanged (`selected_variant=0`, `best_top_gap_score=0`, `failing_profile_count=1`).
|
||||
|
||||
41
docs/sprint261_execution_tracker_2026-02-26.md
Normal file
41
docs/sprint261_execution_tracker_2026-02-26.md
Normal file
@@ -0,0 +1,41 @@
|
||||
# Sprint 261 Execution Tracker - 2026-02-26
|
||||
|
||||
## Scope
|
||||
- `sprint261_plan.md`
|
||||
|
||||
## Implemented
|
||||
|
||||
New tool:
|
||||
- `tools/mcp/synthesize_raw_intrinsic_prompt_pack.py`
|
||||
- Builds a compact intrinsic prompt pack from top-gap signals.
|
||||
- Encodes required ops, executionContract fields, reason keywords, and anti-generic-task constraints.
|
||||
|
||||
Pipeline integration in `tools/mcp/run_sprint_taskitem_pipeline.sh`:
|
||||
- Added:
|
||||
- `WSTONE_NATIVE_RAW_INTRINSIC_PROMPT_PACK`
|
||||
- `WSTONE_NATIVE_RAW_INTRINSIC_PROMPT_MAX_SIGNALS`
|
||||
- Added artifacts:
|
||||
- `01f_raw_intrinsic_prompt_pack_requirements.json`
|
||||
- `01f_raw_intrinsic_prompt_pack_report.json`
|
||||
- Added summary packet:
|
||||
- `native_raw_intrinsic_prompt_pack`
|
||||
|
||||
## Validation Artifacts
|
||||
|
||||
OFF:
|
||||
- `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163844/00_summary.json`
|
||||
|
||||
ON:
|
||||
- `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163855/00_summary.json`
|
||||
- prompt-pack artifacts:
|
||||
- `logs/taskitem_runs/01a_fallback_intake_spec_20260226_163855/01f_raw_intrinsic_prompt_pack_report.json`
|
||||
|
||||
Observed result on hard sample:
|
||||
- prompt pack injected with selected signals (`selected_signal_count=8`)
|
||||
- `selected_variant` remained `0`
|
||||
- `best_top_gap_score` remained `0`
|
||||
- `failing_profile_count` remained `1`
|
||||
|
||||
## Explicit Completion Signal
|
||||
|
||||
- Sprint 261: `DONE` (implemented + validated; intrinsic uplift still pending)
|
||||
6
editor/src/Sprint261IntegrationSummary.h
Normal file
6
editor/src/Sprint261IntegrationSummary.h
Normal file
@@ -0,0 +1,6 @@
|
||||
#pragma once
|
||||
|
||||
// Sprint 261 integration summary:
|
||||
// - Added intrinsic prompt-pack synthesis from top-gap signals.
|
||||
// - Pipeline can inject prompt-pack constraints into normalized requirements pre-generation.
|
||||
// - Summary now includes native_raw_intrinsic_prompt_pack for intrinsic intervention tracing.
|
||||
11
sprint261_plan.md
Normal file
11
sprint261_plan.md
Normal file
@@ -0,0 +1,11 @@
|
||||
# Sprint 261 Plan: Intrinsic Prompt-Pack Intervention
|
||||
|
||||
## Goal
|
||||
Add a stronger model-facing intrinsic prompt pack derived from top-gap signals (schema-style required fields + anti-pattern bans) and inject it before raw generation.
|
||||
|
||||
## Steps
|
||||
- Step 2353: Add prompt-pack synthesis tool from top-gap backlog signals.
|
||||
- Step 2354: Inject prompt-pack normalized requirements in pipeline pre-generation stage.
|
||||
- Step 2355: Emit prompt-pack telemetry in summary packet.
|
||||
- Step 2356: Validate OFF/ON behavior on hard sample.
|
||||
- Step 2357: Add `Sprint261IntegrationSummary.h` and execution tracker.
|
||||
@@ -60,6 +60,8 @@ NATIVE_RAW_TOP_GAP_BACKLOG_FILE="${WSTONE_NATIVE_RAW_TOP_GAP_BACKLOG_FILE:-}"
|
||||
NATIVE_RAW_TOP_GAP_REQUIRE_UPLIFT="${WSTONE_NATIVE_RAW_TOP_GAP_REQUIRE_UPLIFT:-0}"
|
||||
NATIVE_RAW_TOP_GAP_REQUIREMENTS="${WSTONE_NATIVE_RAW_TOP_GAP_REQUIREMENTS:-0}"
|
||||
NATIVE_RAW_TOP_GAP_MAX_SIGNALS="${WSTONE_NATIVE_RAW_TOP_GAP_MAX_SIGNALS:-5}"
|
||||
NATIVE_RAW_INTRINSIC_PROMPT_PACK="${WSTONE_NATIVE_RAW_INTRINSIC_PROMPT_PACK:-0}"
|
||||
NATIVE_RAW_INTRINSIC_PROMPT_MAX_SIGNALS="${WSTONE_NATIVE_RAW_INTRINSIC_PROMPT_MAX_SIGNALS:-8}"
|
||||
NATIVE_RAW_TOP_GAP_ADAPTIVE_RETRY="${WSTONE_NATIVE_RAW_TOP_GAP_ADAPTIVE_RETRY:-0}"
|
||||
NATIVE_RAW_TOP_GAP_RETRY_SIGNALS="${WSTONE_NATIVE_RAW_TOP_GAP_RETRY_SIGNALS:-10}"
|
||||
EXTRA_NORMALIZED_REQUIREMENTS_FILE="${WSTONE_EXTRA_NORMALIZED_REQUIREMENTS_FILE:-}"
|
||||
@@ -127,6 +129,7 @@ NATIVE_RAW_CANDIDATE_SEARCH_JSON='{}'
|
||||
NATIVE_RAW_HARDENING_JSON='{}'
|
||||
NATIVE_RAW_SIGNAL_TARGETED_VARIANTS_JSON='{}'
|
||||
NATIVE_RAW_TOP_GAP_REQUIREMENTS_JSON='{}'
|
||||
NATIVE_RAW_INTRINSIC_PROMPT_PACK_JSON='{}'
|
||||
NATIVE_RAW_ADAPTIVE_RETRY_JSON='{}'
|
||||
INTRINSIC_REQS_JSON='[]'
|
||||
EXTRA_NORMALIZED_REQUIREMENTS_JSON='[]'
|
||||
@@ -359,6 +362,22 @@ if [[ "$NATIVE_RAW_TOP_GAP_REQUIREMENTS" == "1" && -n "$NATIVE_RAW_TOP_GAP_BACKL
|
||||
else
|
||||
NATIVE_RAW_TOP_GAP_REQUIREMENTS_JSON='{"enabled":false}'
|
||||
fi
|
||||
if [[ "$NATIVE_RAW_INTRINSIC_PROMPT_PACK" == "1" && -n "$NATIVE_RAW_TOP_GAP_BACKLOG_FILE" && -f "$NATIVE_RAW_TOP_GAP_BACKLOG_FILE" ]]; then
|
||||
python3 "$ROOT_DIR/tools/mcp/synthesize_raw_intrinsic_prompt_pack.py" \
|
||||
--top-gaps "$NATIVE_RAW_TOP_GAP_BACKLOG_FILE" \
|
||||
--max-signals "$NATIVE_RAW_INTRINSIC_PROMPT_MAX_SIGNALS" \
|
||||
--out "$OUT_DIR/01f_raw_intrinsic_prompt_pack_requirements.json" \
|
||||
--out-report "$OUT_DIR/01f_raw_intrinsic_prompt_pack_report.json" >/dev/null
|
||||
RAW_INTRINSIC_PROMPT_REQS_JSON="$(cat "$OUT_DIR/01f_raw_intrinsic_prompt_pack_requirements.json")"
|
||||
NORMALIZED_REQS="$(jq -nc --argjson base "$NORMALIZED_REQS" --argjson extra "$RAW_INTRINSIC_PROMPT_REQS_JSON" '$base + $extra')"
|
||||
NATIVE_RAW_INTRINSIC_PROMPT_PACK_JSON="$(jq -nc \
|
||||
--argjson enabled true \
|
||||
--arg backlog_file "$NATIVE_RAW_TOP_GAP_BACKLOG_FILE" \
|
||||
--argjson report "$(cat "$OUT_DIR/01f_raw_intrinsic_prompt_pack_report.json")" \
|
||||
'{enabled:$enabled, backlog_file:$backlog_file, report:$report}')"
|
||||
else
|
||||
NATIVE_RAW_INTRINSIC_PROMPT_PACK_JSON='{"enabled":false}'
|
||||
fi
|
||||
CONFLICTS="$(printf '%s' "$INTAKE_JSON" | jq '.conflicts // []')"
|
||||
GEN_ARGS="$(jq -nc --argjson nr "$NORMALIZED_REQS" --argjson cf "$CONFLICTS" --arg strict "$STRICT_EXECUTION_CONTRACT" \
|
||||
'{normalizedRequirements:$nr,conflicts:$cf,strictExecutionContract:($strict == "1")}')"
|
||||
@@ -956,6 +975,7 @@ SUMMARY_JSON="$(jq -nc \
|
||||
--argjson native_profile_autofill "$NATIVE_PROFILE_AUTOFILL_JSON" \
|
||||
--argjson native_intrinsic_boost "$NATIVE_INTRINSIC_BOOST_JSON" \
|
||||
--argjson native_raw_top_gap_requirements "$NATIVE_RAW_TOP_GAP_REQUIREMENTS_JSON" \
|
||||
--argjson native_raw_intrinsic_prompt_pack "$NATIVE_RAW_INTRINSIC_PROMPT_PACK_JSON" \
|
||||
--argjson native_raw_signal_targeted_variants "$NATIVE_RAW_SIGNAL_TARGETED_VARIANTS_JSON" \
|
||||
--argjson native_raw_candidate_search "$NATIVE_RAW_CANDIDATE_SEARCH_JSON" \
|
||||
--argjson native_raw_adaptive_retry "$NATIVE_RAW_ADAPTIVE_RETRY_JSON" \
|
||||
@@ -988,6 +1008,7 @@ SUMMARY_JSON="$(jq -nc \
|
||||
native_profile_autofill: $native_profile_autofill,
|
||||
native_intrinsic_boost: $native_intrinsic_boost,
|
||||
native_raw_top_gap_requirements: $native_raw_top_gap_requirements,
|
||||
native_raw_intrinsic_prompt_pack: $native_raw_intrinsic_prompt_pack,
|
||||
native_raw_signal_targeted_variants: $native_raw_signal_targeted_variants,
|
||||
native_raw_candidate_search: $native_raw_candidate_search,
|
||||
native_raw_adaptive_retry: $native_raw_adaptive_retry,
|
||||
|
||||
80
tools/mcp/synthesize_raw_intrinsic_prompt_pack.py
Normal file
80
tools/mcp/synthesize_raw_intrinsic_prompt_pack.py
Normal file
@@ -0,0 +1,80 @@
|
||||
#!/usr/bin/env python3
|
||||
import argparse
|
||||
import json
|
||||
from pathlib import Path
|
||||
from typing import Dict, List
|
||||
|
||||
|
||||
def load_json(path: Path):
|
||||
with path.open("r", encoding="utf-8") as f:
|
||||
return json.load(f)
|
||||
|
||||
|
||||
def write_json(path: Path, obj) -> None:
|
||||
with path.open("w", encoding="utf-8") as f:
|
||||
json.dump(obj, f, indent=2, sort_keys=True)
|
||||
f.write("\n")
|
||||
|
||||
|
||||
def main() -> int:
|
||||
p = argparse.ArgumentParser(description="Build intrinsic prompt-pack requirements from top-gap missing-signal classes.")
|
||||
p.add_argument("--top-gaps", required=True)
|
||||
p.add_argument("--max-signals", type=int, default=8)
|
||||
p.add_argument("--out", required=True)
|
||||
p.add_argument("--out-report", required=True)
|
||||
args = p.parse_args()
|
||||
|
||||
backlog = load_json(Path(args.top_gaps))
|
||||
prioritized = list(backlog.get("prioritized_missing") or [])[: max(0, args.max_signals)]
|
||||
|
||||
required_ops: List[str] = []
|
||||
required_contracts: List[str] = []
|
||||
required_reasons: List[str] = []
|
||||
for row in prioritized:
|
||||
signal = str(row.get("missing", ""))
|
||||
if signal.startswith("missing_prerequisite_op:"):
|
||||
required_ops.append(signal.split(":", 1)[1])
|
||||
elif signal.startswith("missing_execution_contract:"):
|
||||
required_contracts.append(signal.split(":", 1)[1])
|
||||
elif signal.startswith("missing_reason_keyword:"):
|
||||
required_reasons.append(signal.split(":", 1)[1])
|
||||
|
||||
required_ops = sorted(set(required_ops))
|
||||
required_contracts = sorted(set(required_contracts))
|
||||
required_reasons = sorted(set(required_reasons))
|
||||
|
||||
prompt_text = (
|
||||
"intrinsic raw generation contract: emit concrete taskitems only; "
|
||||
"for each task include taskId/title/prerequisiteOps/reasons/executionContract/dependencyTaskIds; "
|
||||
f"ensure prerequisiteOps include: {', '.join(required_ops) if required_ops else 'none'}; "
|
||||
f"ensure executionContract true fields include: {', '.join(required_contracts) if required_contracts else 'deterministic,rollbackRequired,replayValidationRequired'}; "
|
||||
f"ensure reasons contain keywords: {', '.join(required_reasons) if required_reasons else 'risk,contract,replay'}; "
|
||||
"forbid umbrella tasks like 'implement everything' and require at least one verifiable operation per task."
|
||||
)
|
||||
|
||||
reqs = [
|
||||
{
|
||||
"requirementId": "raw-intrinsic-prompt-pack",
|
||||
"kind": "constraint",
|
||||
"normalizedText": prompt_text,
|
||||
"anchor": "raw_intrinsic_prompt_pack",
|
||||
"sourceLine": 0,
|
||||
"ambiguous": False,
|
||||
}
|
||||
]
|
||||
report = {
|
||||
"status": "ok",
|
||||
"selected_signal_count": len(prioritized),
|
||||
"required_ops": required_ops,
|
||||
"required_contract_fields": required_contracts,
|
||||
"required_reason_keywords": required_reasons,
|
||||
}
|
||||
|
||||
write_json(Path(args.out), reqs)
|
||||
write_json(Path(args.out_report), report)
|
||||
print(json.dumps({"status": "ok", "selected_signal_count": len(prioritized), "out": args.out}, sort_keys=True))
|
||||
return 0
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
raise SystemExit(main())
|
||||
Reference in New Issue
Block a user