Context

The ground truth the platform is written against.

Stages 1–4 of the pipeline build our policy. This directory records everything that constrains it — how the city is elected, who currently governs it, what a municipality is and is not able to do, and what the other factions are promising. It exists for four reasons:

1. The powers baseline. Every policy area’s jurisdiction map starts from the standing one in powers rather than rebuilding it, and adds the area-specific levers.
2. Gap analysis. A proposal is only worth making if it is not already adopted, already promised by five factions, or already tried and dropped. The comparison matrices are how we check.
3. Honest positioning. Where another faction has a good policy, we say so and adopt it. Where ours is genuinely distinct, we should be able to point at the row in the matrix that proves it.
4. Falsifiability. Claims about “nobody is addressing X” are checkable claims. They get sourced like any other.

Structure

context/
├── whose-account.md          The vantage point this platform is written from
├── the-ideological-agenda.md What the city's public argument is actually about
├── what-works.md             The asset base: what is worth building on and not breaking
├── the-election/             The municipal electoral system, the 2024 result, the next election
├── city-hall/                The current council: factions, coalition, the mayor's record
├── powers/                   The standing powers primer: bodies, instruments, constraints
└── matrices/                 Policy comparison matrices, one per policy area

Rules

These carry the same evidence standards as the policy research (see the municipal-policy-methodology skill):

• Everything is dated. Council coalitions shift mid-term, factions split, and officials move. Every factual claim states as of when, and files carry a last-verified date in frontmatter.
• The voting record outranks the platform. Council votes are recorded in the protocols, they are recent, and they concern the actual instruments in play. Where a faction’s votes and its published material conflict, both are shown and the conflict is the finding.
• Positions are quoted, not paraphrased into our framing. If a faction’s position is vague, the matrix cell says “vague” and quotes the vague text — it does not silently steelman or strawman it.
• No score without a criterion. If a matrix rates positions, the rating scale is defined in the matrix file and applied identically across factions, including to our own column.

Our own column

Every matrix includes a column for this project’s proposal, held to the same standard and rated on the same scale — including on deliverability, against the same A/B/C/D classes from the powers doctrine. A comparison matrix in which the author’s own policy wins every row is marketing, not analysis. Where another faction’s instrument is better than ours, the matrix says so and the policy paper is expected to change.