edge-strategy-reviewer

Name: edge-strategy-reviewer
Author: mphinance

15stars3forks

How to install edge-strategy-reviewer?

mphinance/alpha-skills/edge-strategy-reviewer

$npx -y skills add mphinance/alpha-skills --skill edge-strategy-reviewer

Installs into the current project.

Summary

Critically review strategy drafts from edge-strategy-designer for edge plausibility, overfitting risk, sample size adequacy, and execution realism. Use when strategy_drafts/*.yaml exists and needs quality gate before pipeline export. Outputs PASS/REVISE/REJECT verdicts with confi

Files · 1

View on GitHub

SKILL.md

1	# Edge Strategy Reviewer
2
3	Deterministic quality gate for strategy drafts produced by `edge-strategy-designer`.
4
5	## When to Use
6
7	- After `edge-strategy-designer` generates `strategy_drafts/*.yaml`
8	- Before exporting drafts to `edge-candidate-agent` via the pipeline
9	- When manually validating a draft strategy for edge plausibility
10
11	## Prerequisites
12
13	- Strategy draft YAML files (output of `edge-strategy-designer`)
14	- Python 3.10+ with PyYAML
15
16	## Workflow
17
18	1. Load draft YAML files from `--drafts-dir` or a single `--draft` file
19	2. Evaluate each draft against 8 criteria (C1-C8) with weighted scoring
20	3. Compute confidence score (weighted average of all criteria)
21	4. Determine verdict: PASS / REVISE / REJECT
22	5. Assess export eligibility (PASS + export_ready_v1 + exportable family)
23	6. Write review output (YAML or JSON) and optional markdown summary
24
25	## Review Criteria
26
27	\| # \| Criterion \| Weight \| Key Checks \|
28	\|---\|-----------\|--------\|------------\|
29	\| C1 \| Edge Plausibility \| 20 \| Thesis quality, domain terms, mechanism keywords (continuous 50-95) \|
30	\| C2 \| Overfitting Risk \| 20 \| 5-tier filter count scoring (90/80/60/40/10), precise threshold penalty \|
31	\| C3 \| Sample Adequacy \| 15 \| Continuous scoring from estimated annual opportunities (10-95) \|
32	\| C4 \| Regime Dependency \| 10 \| Cross-regime validation \|
33	\| C5 \| Exit Calibration \| 10 \| Stop-loss, reward-to-risk \|
34	\| C6 \| Risk Concentration \| 10 \| Position sizing limits \|
35	\| C7 \| Execution Realism \| 10 \| Volume filter, export consistency \|
36	\| C8 \| Invalidation Quality \| 5 \| Signal count and specificity \|
37
38	## Verdict Logic
39
40	- C1 or C2 severity=fail → immediate REJECT
41	- confidence >= 70, no fail findings → PASS
42	- confidence < 35 → REJECT
43	- Otherwise → REVISE (with revision instructions)
44
45	## Running the Script
46
47	```bash
48	# Review all drafts in a directory
49	python3 skills/edge-strategy-reviewer/scripts/review_strategy_drafts.py \
50	--drafts-dir reports/edge_strategy_drafts/ \
51	--output-dir reports/
52
53	# Single draft review
54	python3 skills/edge-strategy-reviewer/scripts/review_strategy_drafts.py \
55	--draft reports/edge_strategy_drafts/draft_xxx.yaml \
56	--output-dir reports/
57
58	# JSON output with markdown summary
59	python3 skills/edge-strategy-reviewer/scripts/review_strategy_drafts.py \
60	--drafts-dir reports/edge_strategy_drafts/ \
61	--output-dir reports/ \
62	--format json \
63	--markdown-summary
64
65	# Strict export mode: export-eligible drafts with any warn → REVISE
66	python3 skills/edge-strategy-reviewer/scripts/review_strategy_drafts.py \
67	--drafts-dir reports/edge_strategy_drafts/ \
68	--output-dir reports/ \
69	--strict-export
70	```
71
72	## Output Format
73
74	Primary output: `review.yaml` (or `review.json`)
75
76	```yaml
77	generated_at_utc: "2026-02-28T12:00:00+00:00"
78	source:
79	drafts_dir: "/path/to/strategy_drafts"
80	draft_count: 4
81	summary:
82	total: 4
83	PASS: 1
84	REVISE: 2
85	REJECT: 1
86	export_eligible: 1
87	reviews:
88	- draft_id: "draft_xxx_core"
89	verdict: "PASS"
90	confidence_score: 80
91	export_eligible: true
92	findings: [...]
93	revision_instructions: []
94	```
95
96	## Resources
97
98	- `references/review_criteria.md` — Detailed scoring rubric for C1-C8
99	- `references/overfitting_checklist.md` — Overfitting detection heuristics

People who liked this also viewed

6 picks

Type

edge-strategy-reviewer

15stars3forks

View on GitHub

How to install edge-strategy-reviewer?

mphinance/alpha-skills/edge-strategy-reviewer

$npx -y skills add mphinance/alpha-skills --skill edge-strategy-reviewer

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Edge Strategy Reviewer
2
3	Deterministic quality gate for strategy drafts produced by `edge-strategy-designer`.
4
5	## When to Use
6
7	- After `edge-strategy-designer` generates `strategy_drafts/*.yaml`
8	- Before exporting drafts to `edge-candidate-agent` via the pipeline
9	- When manually validating a draft strategy for edge plausibility
10
11	## Prerequisites
12
13	- Strategy draft YAML files (output of `edge-strategy-designer`)
14	- Python 3.10+ with PyYAML
15
16	## Workflow
17
18	1. Load draft YAML files from `--drafts-dir` or a single `--draft` file
19	2. Evaluate each draft against 8 criteria (C1-C8) with weighted scoring
20	3. Compute confidence score (weighted average of all criteria)
21	4. Determine verdict: PASS / REVISE / REJECT
22	5. Assess export eligibility (PASS + export_ready_v1 + exportable family)
23	6. Write review output (YAML or JSON) and optional markdown summary
24
25	## Review Criteria
26
27	\| # \| Criterion \| Weight \| Key Checks \|
28	\|---\|-----------\|--------\|------------\|
29	\| C1 \| Edge Plausibility \| 20 \| Thesis quality, domain terms, mechanism keywords (continuous 50-95) \|
30	\| C2 \| Overfitting Risk \| 20 \| 5-tier filter count scoring (90/80/60/40/10), precise threshold penalty \|
31	\| C3 \| Sample Adequacy \| 15 \| Continuous scoring from estimated annual opportunities (10-95) \|
32	\| C4 \| Regime Dependency \| 10 \| Cross-regime validation \|
33	\| C5 \| Exit Calibration \| 10 \| Stop-loss, reward-to-risk \|
34	\| C6 \| Risk Concentration \| 10 \| Position sizing limits \|
35	\| C7 \| Execution Realism \| 10 \| Volume filter, export consistency \|
36	\| C8 \| Invalidation Quality \| 5 \| Signal count and specificity \|
37
38	## Verdict Logic
39
40	- C1 or C2 severity=fail → immediate REJECT
41	- confidence >= 70, no fail findings → PASS
42	- confidence < 35 → REJECT
43	- Otherwise → REVISE (with revision instructions)
44
45	## Running the Script
46
47	```bash
48	# Review all drafts in a directory
49	python3 skills/edge-strategy-reviewer/scripts/review_strategy_drafts.py \
50	--drafts-dir reports/edge_strategy_drafts/ \
51	--output-dir reports/
52
53	# Single draft review
54	python3 skills/edge-strategy-reviewer/scripts/review_strategy_drafts.py \
55	--draft reports/edge_strategy_drafts/draft_xxx.yaml \
56	--output-dir reports/
57
58	# JSON output with markdown summary
59	python3 skills/edge-strategy-reviewer/scripts/review_strategy_drafts.py \
60	--drafts-dir reports/edge_strategy_drafts/ \
61	--output-dir reports/ \
62	--format json \
63	--markdown-summary
64
65	# Strict export mode: export-eligible drafts with any warn → REVISE
66	python3 skills/edge-strategy-reviewer/scripts/review_strategy_drafts.py \
67	--drafts-dir reports/edge_strategy_drafts/ \
68	--output-dir reports/ \
69	--strict-export
70	```
71
72	## Output Format
73
74	Primary output: `review.yaml` (or `review.json`)
75
76	```yaml
77	generated_at_utc: "2026-02-28T12:00:00+00:00"
78	source:
79	drafts_dir: "/path/to/strategy_drafts"
80	draft_count: 4
81	summary:
82	total: 4
83	PASS: 1
84	REVISE: 2
85	REJECT: 1
86	export_eligible: 1
87	reviews:
88	- draft_id: "draft_xxx_core"
89	verdict: "PASS"
90	confidence_score: 80
91	export_eligible: true
92	findings: [...]
93	revision_instructions: []
94	```
95
96	## Resources
97
98	- `references/review_criteria.md` — Detailed scoring rubric for C1-C8
99	- `references/overfitting_checklist.md` — Overfitting detection heuristics

People who liked this also viewed

6 picks

Type