policyengine-prior-scores

Name: policyengine-prior-scores
Author: PolicyEngine

31stars5forks

How to install policyengine-prior-scores?

policyengine/policyengine-claude/policyengine-prior-scores

$npx -y skills add PolicyEngine/policyengine-claude --skill policyengine-prior-scores

Installs into the current project.

Summary

Curated anchor list of PolicyEngine's previously-published scored reforms, plus the map of the real prior-scores infrastructure in this repo (presets, the scorekeepers registry, the /prior-scores command, and the prior-scores-finder agent). Use to find a benchmark magnitude to an

Files · 1

View on GitHub

SKILL.md

1	# PolicyEngine prior scores
2
3	Finding a prior score to anchor a new analysis has two parts: a curated anchor list of
4	notable PE reforms (below, with URLs) and the real infrastructure in this repo that
5	enumerates presets, external scorekeepers, and local archived runs. Earlier versions of this
6	skill described a `data/prior-scores.json` file with `get_priors_by_program()` query functions —
7	those never existed. The sections below point only at things that are actually on disk or live.
8
9	## When to use
10
11	- `/analyze-policy` Stage 3 (find prior scores) and Stage 5 (compare the microsim to them).
12	- The `/prior-scores` command (external-benchmark research without running the microsim).
13	- The `prior-scores-finder` agent's Tier-1 lookup.
14	- Writing a research post and citing PE's prior work on a similar reform.
15
16	## Real infrastructure in this repo
17
18	### Presets — named reform-dicts with published scores
19
20	`presets/reforms/.yaml` and `presets/baselines/.yaml` are callable-by-name reform/baseline
21	dicts. A preset may carry a `published_scores` block recording what external sources scored that
22	exact shape — which is what makes external corroboration turn-key.
23
24	```bash
25	python3 scripts/presets.py list # all presets
26	python3 scripts/presets.py list --category reform --country us
27	python3 scripts/presets.py show arpa-ctc-restoration # prints the reform_dict + published_scores
28	```
29
30	```python
31	from scripts.presets import load_preset
32	preset = load_preset("arpa-ctc-restoration")
33	preset["reform_dict"] # ready to pass as a reform
34	preset["published_scores"] # [{source: JCT, ten_year_billion: 1100, ...}, ...]
35	```
36
37	Presets on disk today (verify with `python3 scripts/presets.py list` — `load_preset` raises
38	on names that don't exist): reforms `arpa-ctc-restoration`, `obbba-salt-bump`; baseline
39	`tcja-extension`. A `pre-obbba` baseline is wanted but not yet authored. Add a preset by copying
40	a nearby file and, if any external source scored the shape, filling in `published_scores` — see
41	`presets/README.md`.
42
43	### Scorekeepers — the external-benchmark registry
44
45	`presets/scorekeepers.yaml` enumerates the official fiscal offices (Tier 2) and think-tanks
46	(Tier 3) to consult per jurisdiction and domain — so the search casts a wide, non-US-only net
47	without hardcoding sources into agents. Add a scorekeeper by appending to the YAML; no code
48	change.
49
50	```bash
51	python3 scripts/scorekeepers.py list --country us --tier 2 # JCT, CBO, OTA, SSA-OCACT
52	python3 scripts/scorekeepers.py list --country us --tier 3 # CRFB, TPC, Tax Foundation, CBPP, ITEP, ...
53	python3 scripts/scorekeepers.py list --country uk --tier 3 # IFS, Resolution Foundation, IPPR, NIESR, ...
54	python3 scripts/scorekeepers.py list --country ca # PBO, DoF-Canada, CD Howe, IRPP, ...
55	```
56
57	```python
58	from scripts.scorekeepers import list_scorekeepers
59	tier_2 = list_scorekeepers(country="us", tier=2)
60	```
61
62	Each entry has `search_hints` (e.g. `site:crfb.org`) for constructing WebSearch queries, plus
63	`domains` and `caveats`.
64
65	### Command and agent
66
67	- `/prior-scores` (`targets/claude/commands/prior-scores.md`) — standalone external-benchmark
68	research: loads the scorekeepers registry filtered by country/tier/domain, also consults the
69	local `analyses/` archive (Tier 0, via `scripts/analyses_kb.py`) and PE published research
70	(Tier 1, this skill), and returns a ranked benchmark cluster. It does NOT run the microsim.
71	- `prior-scores-finder` (`targets/claude/agents/prior-scores-finder.md`) — the agent behind
72	Stage 3. It runs BLIND to the current run's own microsim result (pre-registration), records
73	each source with its framing at registration time, and requires all tiers: Tier 0 (local
74	archive), Tier 1 (PE published — this skill), Tier 2 (official), Tier 3 (think-tanks). A PE
75	prior does not excuse skipping the external tiers.
76
77	## Curated anchor reforms
78
79	Notable PE-published scores, kept as quick anchors. Verify the live number before quoting — these
80	are pointers, not a maintained datab

People who liked this also viewed

6 picks

Type

policyengine-prior-scores

31stars5forks

View on GitHub

How to install policyengine-prior-scores?

policyengine/policyengine-claude/policyengine-prior-scores

$npx -y skills add PolicyEngine/policyengine-claude --skill policyengine-prior-scores

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# PolicyEngine prior scores
2
3	Finding a prior score to anchor a new analysis has two parts: a curated anchor list of
4	notable PE reforms (below, with URLs) and the real infrastructure in this repo that
5	enumerates presets, external scorekeepers, and local archived runs. Earlier versions of this
6	skill described a `data/prior-scores.json` file with `get_priors_by_program()` query functions —
7	those never existed. The sections below point only at things that are actually on disk or live.
8
9	## When to use
10
11	- `/analyze-policy` Stage 3 (find prior scores) and Stage 5 (compare the microsim to them).
12	- The `/prior-scores` command (external-benchmark research without running the microsim).
13	- The `prior-scores-finder` agent's Tier-1 lookup.
14	- Writing a research post and citing PE's prior work on a similar reform.
15
16	## Real infrastructure in this repo
17
18	### Presets — named reform-dicts with published scores
19
20	`presets/reforms/.yaml` and `presets/baselines/.yaml` are callable-by-name reform/baseline
21	dicts. A preset may carry a `published_scores` block recording what external sources scored that
22	exact shape — which is what makes external corroboration turn-key.
23
24	```bash
25	python3 scripts/presets.py list # all presets
26	python3 scripts/presets.py list --category reform --country us
27	python3 scripts/presets.py show arpa-ctc-restoration # prints the reform_dict + published_scores
28	```
29
30	```python
31	from scripts.presets import load_preset
32	preset = load_preset("arpa-ctc-restoration")
33	preset["reform_dict"] # ready to pass as a reform
34	preset["published_scores"] # [{source: JCT, ten_year_billion: 1100, ...}, ...]
35	```
36
37	Presets on disk today (verify with `python3 scripts/presets.py list` — `load_preset` raises
38	on names that don't exist): reforms `arpa-ctc-restoration`, `obbba-salt-bump`; baseline
39	`tcja-extension`. A `pre-obbba` baseline is wanted but not yet authored. Add a preset by copying
40	a nearby file and, if any external source scored the shape, filling in `published_scores` — see
41	`presets/README.md`.
42
43	### Scorekeepers — the external-benchmark registry
44
45	`presets/scorekeepers.yaml` enumerates the official fiscal offices (Tier 2) and think-tanks
46	(Tier 3) to consult per jurisdiction and domain — so the search casts a wide, non-US-only net
47	without hardcoding sources into agents. Add a scorekeeper by appending to the YAML; no code
48	change.
49
50	```bash
51	python3 scripts/scorekeepers.py list --country us --tier 2 # JCT, CBO, OTA, SSA-OCACT
52	python3 scripts/scorekeepers.py list --country us --tier 3 # CRFB, TPC, Tax Foundation, CBPP, ITEP, ...
53	python3 scripts/scorekeepers.py list --country uk --tier 3 # IFS, Resolution Foundation, IPPR, NIESR, ...
54	python3 scripts/scorekeepers.py list --country ca # PBO, DoF-Canada, CD Howe, IRPP, ...
55	```
56
57	```python
58	from scripts.scorekeepers import list_scorekeepers
59	tier_2 = list_scorekeepers(country="us", tier=2)
60	```
61
62	Each entry has `search_hints` (e.g. `site:crfb.org`) for constructing WebSearch queries, plus
63	`domains` and `caveats`.
64
65	### Command and agent
66
67	- `/prior-scores` (`targets/claude/commands/prior-scores.md`) — standalone external-benchmark
68	research: loads the scorekeepers registry filtered by country/tier/domain, also consults the
69	local `analyses/` archive (Tier 0, via `scripts/analyses_kb.py`) and PE published research
70	(Tier 1, this skill), and returns a ranked benchmark cluster. It does NOT run the microsim.
71	- `prior-scores-finder` (`targets/claude/agents/prior-scores-finder.md`) — the agent behind
72	Stage 3. It runs BLIND to the current run's own microsim result (pre-registration), records
73	each source with its framing at registration time, and requires all tiers: Tier 0 (local
74	archive), Tier 1 (PE published — this skill), Tier 2 (official), Tier 3 (think-tanks). A PE
75	prior does not excuse skipping the external tiers.
76
77	## Curated anchor reforms
78
79	Notable PE-published scores, kept as quick anchors. Verify the live number before quoting — these
80	are pointers, not a maintained datab

People who liked this also viewed

6 picks

Type