Skip to content

feat: domain-aware normalizer for assumption validation (Phase 2)#72

Closed
82deutschmark wants to merge 5 commits into
PlanExeOrg:mainfrom
VoynichLabs:feature/domain-normalizer
Closed

feat: domain-aware normalizer for assumption validation (Phase 2)#72
82deutschmark wants to merge 5 commits into
PlanExeOrg:mainfrom
VoynichLabs:feature/domain-normalizer

Conversation

@82deutschmark

@82deutschmark 82deutschmark commented Feb 25, 2026

Copy link
Copy Markdown
Collaborator

Phase 2: Domain-Aware Assumption Normalizer

What this does

Adds a DomainNormalizer that auto-detects the domain of a plan (carpenter, dentist, personal, startup, non-profit) and normalizes QuantifiedAssumptions accordingly:

  • Currency → domain default (DKK for carpenter, USD for dentist) + EUR equivalent
  • Units → metric internally
  • Confidence keywords → domain-aware ("I've done this 50 times" = high for carpenter)

Why it matters

PlanExe's role in 2026 is as a trusted auditing layer for autonomous agents — not just plan generation. Agents run in bubbles and hallucinate assumptions. This normalizer ensures FermiSanityCheck speaks the right language for each domain before flagging assumptions as sane or suspect.

Test results

11/11 tests pass (8 normalizer + 3 FermiSanityCheck).

DAG placement

MakeAssumptions → FermiSanityCheck → DomainNormalizer → DistillAssumptions

Related

EgonBot and others added 5 commits February 25, 2026 15:48
docs: add proposal triage summary
1e824bf
feat: add quantified assumption extractor
d5a9bdd
docs: add quantified assumption schema reference
9cc1e91
feat: integrate fermi sanity check
e2c6a27
feat: Add domain-aware normalizer for FermiSanityCheck
dbc70f1 
- Loads domain profiles (Carpenter, Dentist, Personal) from YAML
- Auto-detects domain from assumption signals (currency, units, keywords)
- Normalizes currency to domain default + EUR equivalent
- Normalizes units to metric (with conversion tables)
- Re-assesses confidence per domain keywords
- Batch normalization support
- Unit tests cover detection, normalization, conversions

Addresses Simon's feedback on hardcoded lists + Mark's requirement for
clean, domain-aware outputs for AI agents.
@82deutschmark 82deutschmark mentioned this pull request Feb 25, 2026
Merged
@82deutschmark

82deutschmark commented Mar 6, 2026

Copy link
Copy Markdown
Collaborator Author

Closing this PR per Simon's review. The approach has fundamental issues that need proper architecture first:

  1. Hardcoded currency rates — currency conversion needs to be a proper service with live/cached exchange rates, not hardcoded values
  2. Hardcoded CURRENCY_UNITS — PlanExe works with all currencies the LLM knows about, not a handful
  3. Fragile path handling — DOMAIN_PROFILES_PATH breaks if files move
  4. Uppercase LOGGER — inconsistent with Python conventions

This feature is being split into prerequisite components, each as a separate PR:

  • Currency exchange rate service (proposal PR incoming)
  • Resource/cost data service (proposal PR incoming)
  • Domain profile system (future PR)
  • Quantified assumptions extractor (future PR)
  • Domain-aware normalizer (depends on all above)

Each component will be submitted as a focused, reviewable PR. The proposals will be submitted to docs/proposals/ for review before implementation.

82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: currency exchange rate service architecture
c9aa90d 
Analyzes three approaches (Python module, Docker service, MCP tool) with
pros/cons for each. Recommends starting with a Python module with local
cache, evolving to hybrid module+MCP. Replaces hardcoded rates from PR PlanExeOrg#72.
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: location-aware resource and cost data service
c186e78 
Designs a tiered data architecture for real-world cost lookups:
- Tier 1: Authoritative sources (ILO, World Bank, BLS)
- Tier 2: Interpolation (geographic, PPP adjustment, economic similarity)
- Tier 3: LLM-assisted estimation with confidence scoring

Includes data model, interpolation strategy, phased implementation plan,
and integration points with PlanExe pipeline. Replaces hardcoded cost
assumptions from PR PlanExeOrg#72.
@82deutschmark 82deutschmark mentioned this pull request Mar 6, 2026
Merged
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: currency exchange rate service architecture
f883794 
Analyzes three approaches (Python module, Docker service, MCP tool) with
pros/cons for each. Recommends starting with a Python module with local
cache, evolving to hybrid module+MCP. Replaces hardcoded rates from PR PlanExeOrg#72.
@82deutschmark 82deutschmark mentioned this pull request Mar 6, 2026
Merged
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: domain-aware assumption normalizer (captures PR PlanExeOrg#72…
7db9ba7 
… intent)

Documents the original vision of PR PlanExeOrg#72 (domain detection, currency
normalization, unit conversion, confidence calibration, Fermi sanity
checks), redesigned to build on the prerequisite Currency Service
(PR PlanExeOrg#147) and Resource/Cost Data (PR PlanExeOrg#146) proposals.

Split into 4 implementation phases, each as a separate future PR.
Addresses all review feedback from PR PlanExeOrg#72 (no hardcoded rates, no
fragile paths, standard logger convention, configurable profiles).
@82deutschmark 82deutschmark mentioned this pull request Mar 6, 2026
Merged
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: domain-aware assumption normalizer (captures PR PlanExeOrg#72…
a20258c 
… intent)

Comprehensive design document covering:
- Domain detection and assumption extraction
- Fermi sanity checking against real-world benchmarks
- Currency/unit normalization via Currency Service
- Confidence calibration per domain
- Ethical framework: fair wage floors, workplace safety standards,
  child labor detection, working conditions assessment

Built on Currency Service (PR PlanExeOrg#147) and Resource/Cost Data (PR PlanExeOrg#146).
Split into 4 phased implementation PRs. Addresses all PR PlanExeOrg#72 feedback.
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: domain-aware assumption normalizer (captures PR PlanExeOrg#72…
27b2df9 
… intent)

Comprehensive design document covering:
- Domain detection and assumption extraction
- Fermi sanity checking against real-world benchmarks
- Currency/unit normalization via Currency Service
- Confidence calibration per domain
- Ethical framework: fair wage floors, workplace safety standards,
  child labor detection, working conditions assessment

Built on Currency Service (PR PlanExeOrg#147) and Resource/Cost Data (PR PlanExeOrg#146).
Split into 4 phased implementation PRs. Addresses all PR PlanExeOrg#72 feedback.
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: domain-aware assumption normalizer (captures PR PlanExeOrg#72…
94dff9e 
… intent)

Comprehensive design document covering:
- Domain detection and assumption extraction
- Fermi sanity checking against real-world benchmarks
- Currency/unit normalization via Currency Service
- Confidence calibration per domain
- Ethical framework: fair wage floors, workplace safety standards,
  child labor detection, working conditions assessment

Built on Currency Service (PR PlanExeOrg#147) and Resource/Cost Data (PR PlanExeOrg#146).
Split into 4 phased implementation PRs. Addresses all PR PlanExeOrg#72 feedback.
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: domain-aware assumption normalizer (captures PR PlanExeOrg#72…
dc8838e 
… intent)

Comprehensive design document covering:
- Domain detection and assumption extraction
- Fermi sanity checking against real-world benchmarks
- Currency/unit normalization via Currency Service
- Confidence calibration per domain
- Ethical framework: fair wage floors, workplace safety standards,
  child labor detection, working conditions assessment

Built on Currency Service (PR PlanExeOrg#147) and Resource/Cost Data (PR PlanExeOrg#146).
Split into 4 phased implementation PRs. Addresses all PR PlanExeOrg#72 feedback.
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: domain-aware assumption normalizer (captures PR PlanExeOrg#72…
8f4cf02 
… intent)

Comprehensive design document covering:
- Domain detection and assumption extraction
- Fermi sanity checking against real-world benchmarks
- Currency/unit normalization via Currency Service
- Confidence calibration per domain
- Ethical framework: fair wage floors, workplace safety standards,
  child labor detection, working conditions assessment

Built on Currency Service (PR PlanExeOrg#147) and Resource/Cost Data (PR PlanExeOrg#146).
Split into 4 phased implementation PRs. Addresses all PR PlanExeOrg#72 feedback.
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: domain-aware assumption normalizer (captures PR PlanExeOrg#72…
d67f534 
… intent)

Comprehensive design document covering:
- Domain detection and assumption extraction
- Fermi sanity checking against real-world benchmarks
- Currency/unit normalization via Currency Service
- Confidence calibration per domain
- Ethical framework: fair wage floors, workplace safety standards,
  child labor detection, working conditions assessment

Built on Currency Service (PR PlanExeOrg#147) and Resource/Cost Data (PR PlanExeOrg#146).
Split into 4 phased implementation PRs. Addresses all PR PlanExeOrg#72 feedback.
82deutschmark added a commit to VoynichLabs/PlanExe2026 that referenced this pull request Mar 6, 2026
@82deutschmark
proposal: domain-aware assumption normalizer (captures PR PlanExeOrg#72…
6a51c7a 
… intent)

Comprehensive design document covering:
- Domain detection and assumption extraction
- Fermi sanity checking against real-world benchmarks
- Currency/unit normalization via Currency Service
- Confidence calibration per domain
- Ethical framework: fair wage floors, workplace safety standards,
  child labor detection, working conditions assessment

Built on Currency Service (PR PlanExeOrg#147) and Resource/Cost Data (PR PlanExeOrg#146).
Split into 4 phased implementation PRs. Addresses all PR PlanExeOrg#72 feedback.
neoneye added a commit that referenced this pull request Mar 6, 2026
@neoneye
Merge pull request #148 from VoynichLabs/proposal/domain-aware-normal…
3537a66 
…izer

proposal: Domain-aware assumption normalizer (captures PR #72 intent)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@82deutschmark