Which plan do I need to run one?

Team. Both sides read a tender document you upload, and document grounding starts at Team — the requirement text is quoted verbatim from your file and anchored to the paragraph it came from. The pricing page carries the run allowances.

Bidder side or buyer side — which one do we run?

Run the review if you are bidding: it assesses whether your posture covers each requirement, lot by lot, and scores coverage against a threshold you set. Run the audit if you are the contracting authority: it tests your own requirements for proportionality, non-discrimination and transparency, and flags a selection criterion doing award work. Same decomposition, same evidence trail, opposite question.

Does it decide whether to bid, or whether a requirement is unlawful?

No. Both sides produce findings with a status and a citation, for a bid/no-bid meeting or a legal review. The audit rates each defect for challengeability rather than declaring a requirement unlawful, and Ansvar gives no legal advice.

What grounds a finding?

The requirement text is a verbatim quote from your uploaded tender, cited to its own paragraph segment with a content hash. The verdict is then enriched across the primary regime, the horizontal ones, the sector regulator, case law and authority guidance before it is recorded — so a reviewer can trace both what the tender said and what the verdict rests on.

What if a requirement has no citation you serve?

The finding records the basis as unresolved, and the report's quality warnings say how many did. Thin evidence and sub-threshold coverage are reported on the face of the document rather than smoothed over — a review that hides its own weak spots is worth less than one that names them.

Workflow · Tender review

Public tender review & audit

Two sides of the same tender: whether a bid covers what the tender demands, and whether the tender's own requirements are lawful.

Ansvar is a gateway for the AI assistant your team already uses — Claude, Microsoft Copilot, any MCP client. Connect it and your agent runs the workflow; the server enforces the stages and fetches every citation.

Compare plans

Per requirementone assessment step each, not one pass over the document

Paragraph-anchoredevery requirement cited to its own segment, with a content hash

Sweden · NetherlandsLOU and Aanbestedingswet overlays on the shared spine

Both sides start by decomposing the tender into lots and requirements, each quoted verbatim from the document you upload and anchored to its own paragraph with a content hash — so a reviewer can check the quote, not just trust it. From there the question splits. Bidding, the run assesses each requirement against your posture and scores coverage per lot against a threshold you set. Contracting, it audits your own requirements: proportionality, non-discrimination, transparency, whether a selection criterion is doing award work, whether an exclusion ground cites the right basis, and it rates each defect for challengeability. The Swedish and Dutch variants overlay LOU and the Aanbestedingswet on the same spine.

the family

Two sides of the same tender

Public Tender Review

Bidder side: does your posture cover every requirement, lot by lot.

Public Tender Review — Netherlands (Classical, Aw 2012)

The bid review under the Dutch classical regime, Aanbestedingswet 2012.

Public Tender Review — Sweden (Classical, LOU)

The bid review under the Swedish classical regime, LOU.

Public Tender Audit (Buyer-Side Lawfulness Review)

Buyer side: are your own requirements proportionate, non-discriminatory and transparent.

Public Tender Audit — Netherlands (Classical, Aw 2012)

The buyer-side audit under the Dutch regime, with PIANOo as reference.

Public Tender Audit — Sweden (Classical, LOU)

The buyer-side audit under LOU, with Upphandlingsmyndigheten as reference.

Each requirement runs the full enrichment pass before a verdict is recorded: the primary regime, the horizontal ones, the sector regulator, case law, and authority guidance. Where none of them grounds a requirement, the finding records the basis as unresolved rather than inheriting an invented one.

how a run works

Four stages the server enforces

1
Decomposition & scoping
the tender broken into lots and requirements, each quoted verbatim and anchored to its paragraph in your upload
2
Per-requirement assessment
one step per requirement: coverage against your posture on the bid side, lawfulness on the buyer side
3
Findings review
the findings before the report, with the coverage or defect rate measured against the threshold you configured
4
Report
requirement-by-requirement findings — structured for your agent, rendered for your auditor

ask your agent

Paste one of these to start

Using Ansvar, review our bid against this tender requirement by requirement, and cite every requirement to its paragraph in the tender document.

Using Ansvar, audit our own draft tender for lawfulness: proportionality, non-discrimination, transparency, and whether any selection criterion is doing award work.

install the skill

A prompt starts one run. The skill is the same guidance installed once — the run loop, the evidence and citation rules, and the starters — so your agent works this way in every conversation, not only the ones you remember to paste into. Install it as a skill in Claude or Claude Code, or paste the same file into Microsoft Copilot or a custom GPT's instructions.

Tender Review and Audit · version 1.0.4 · SKILL.md · ZIP · how to install it

The Tender review family runs on Team and above — run allowances and what each tier adds live on the pricing page.

Questions buyers ask first

Which plan do I need to run one?: Team. Both sides read a tender document you upload, and document grounding starts at Team — the requirement text is quoted verbatim from your file and anchored to the paragraph it came from. The pricing page carries the run allowances.
Bidder side or buyer side — which one do we run?: Run the review if you are bidding: it assesses whether your posture covers each requirement, lot by lot, and scores coverage against a threshold you set. Run the audit if you are the contracting authority: it tests your own requirements for proportionality, non-discrimination and transparency, and flags a selection criterion doing award work. Same decomposition, same evidence trail, opposite question.
Does it decide whether to bid, or whether a requirement is unlawful?: No. Both sides produce findings with a status and a citation, for a bid/no-bid meeting or a legal review. The audit rates each defect for challengeability rather than declaring a requirement unlawful, and Ansvar gives no legal advice.
What grounds a finding?: The requirement text is a verbatim quote from your uploaded tender, cited to its own paragraph segment with a content hash. The verdict is then enriched across the primary regime, the horizontal ones, the sector regulator, case law and authority guidance before it is recorded — so a reviewer can trace both what the tender said and what the verdict rests on.
What if a requirement has no citation you serve?: The finding records the basis as unresolved, and the report's quality warnings say how many did. Thin evidence and sub-threshold coverage are reported on the face of the document rather than smoothed over — a review that hides its own weak spots is worth less than one that names them.

as a service

Prefer we run it?

Every workflow here is also an expert-run service: we run it against your systems, review the output as practitioners, and hand over the finished deliverable. See the services page for how engagements work, or contact us to scope one.

Run it against your own systems

Connect the AI client you already use and ask your first cited question — Free, Solo, Premium and Team are self-serve.

Setup guide