ai-agent-contract-language-pack

name

ai-agent-contract-language-pack

description

Use when the AI-agent proposal must drop contract-grade exhibit language into the financial proposal, the MSA, or the SoW. Curates the ready-to-paste exhibits — pricing exhibit, SLA exhibit, credit and refund exhibit, abandonment-and-refund policy, vendor-cost-pass-through, FX corridor, dispute resolution, audit rights for SLA metrics, action-accountability — and tells the writer which to deploy for which engagement shape.

AI-Agent Contract Language Pack

Acknowledgement: Shared by Peter Bamuhigire, techguypeter.com, [phone].

Use When

• The proposal stage has moved from value narrative to contractual language.
• Procurement, legal, or finance has asked for the actual clauses, not paragraph descriptions.
• The bid must include exhibit-grade language as annexes (Pricing Exhibit, SLA Exhibit, Refund Exhibit).
• The agency's legal team is preparing the MSA / SoW and needs a starting position.
• A competitive RFP requires drop-in commercial language.

Do Not Use When

• The proposal is still at value-narrative stage and clauses would over-commit the agency (use ai-agent-pricing-and-packaging-proposal for paragraph-grade language).
• The engagement is non-agent (use saas-msa-dpa-sla-template-language).

Required Inputs

• The chosen pricing pattern and packaging shape.
• The SLA class and credit schedule.
• The autonomy-ramp curve.
• The kill-switch architecture and drill log.
• The audit-log completeness number and retention.
• The model-provider list and pass-through terms.
• The FX baseline and corridor.
• The buyer's governing-law and dispute-resolution preference.
• The agency's legal team's approved starting positions on liability, indemnity, insurance.

Workflow

1. Confirm the pricing pattern (ai-agent-pricing-and-packaging-proposal) and packaging shape (ai-agent-commercial-packaging). The pack adapts to both.
2. Confirm the SLA class (ai-agent-sla-and-credit-schedule).
3. Assemble the contract pack from the references below — pricing exhibit + SLA exhibit + credit/refund exhibit + MSA addendum + abandonment-and-refund policy + dispute resolution. Each is drop-in.
4. Tailor named placeholders — currency, jurisdiction, reference rate, model providers, autonomy ramp, kill-switch time, intervention rate ceiling, audit-log completeness, retention period.
5. Run the internal consistency pass — pricing, SLA, refund, MSA addendum all reference the same numbers, the same definitions, and the same evidence trail.
6. Hand the pack to the agency's legal team for sign-off before submission.
7. Output the Contract Language Annex for the proposal and a parallel MSA / SoW Insertion Pack for the legal team.

The Contract Language Pack — Index

Exhibit	Source reference	Purpose
Pricing Exhibit	../references/ai-agent-pricing-exhibit-template.md	Defines unit; states price per unit; included volumes; overage rates; floor; intervention credit; vendor pass-through; FX; fair-use.
SLA Exhibit	../references/ai-agent-sla-exhibit-template.md	Defines the SLA class; commits the four agent metrics and the three guardrail SLAs; credit formulas; cap; out-clauses; reporting; termination right.
Credit and Refund Exhibit	../references/ai-agent-credit-and-refund-clauses.md	Drop-in credit clauses (per-metric and per-event); refund clauses (irreversible-action, intervention overshoot, regulator action).
Abandonment-and-Refund Policy	../references/ai-agent-abandonment-and-refund-policy-template.md	Customer-facing policy on when the agency will abandon the deployment and refund.
MSA Addendum	../references/ai-agent-msa-addendum-template.md	Drop-in MSA addendum extending the SaaS MSA with action accountability, kill-switch SLA, replay availability, fee-for-evidence-pack on dispute, sub-cap for irreversible-action liability.
Dispute Resolution and Audit Rights	../references/ai-agent-dispute-resolution-and-audit-rights.md	SLA dispute mechanics; audit-log audit rights; evidence-pack fee; escalation path.
Outcome-Pricing Exhibits	../references/ai-agent-outcome-pricing-clauses.md	Gain-share, success-fee, hybrid base-plus-success language.
Vendor Cost Pass-Through	../references/ai-agent-vendor-cost-pass-through.md	Pass-through formula, indexed pricing, sovereign-AI handling.
Success-Definition Language	../references/ai-agent-success-definition-language.md	How to define a resolution / outcome / action with counter-example rule and cooling-off.
Renewal and True-Up Clauses	../references/ai-agent-renewal-and-true-up-clauses.md	Renewal mechanics; true-up; ramp-down protection; autonomy-progression price-step.

Trade-Not-Give Discipline

Every exhibit is drafted with a default position. The agency does not give away a default without trading it for something:

• Buyer wants a longer audit-log retention → agency trades against a fee-for-evidence-pack on dispute.
• Buyer wants higher SLA class → agency trades against package shape (move from Included to Add-on; from Add-on to Standalone).
• Buyer wants tighter intervention credit → agency trades against a longer cooling-off period in the outcome definition.
• Buyer wants uncapped credits → agency declines; credits are the sole remedy except where contract or law requires otherwise.
• Buyer wants no pass-through → agency declines; pass-through is the agency's protection against model-price shock.

Quality Standards

• Every exhibit is internally consistent — same definitions, same numbers, same evidence trail.
• Every placeholder is named and filled.
• No "TBD" in submission-grade language.
• Pricing, SLA, credit, refund, MSA addendum reference the same autonomy ramp, the same intervention rate ceiling, the same audit-log completeness, the same retention.
• Legal team has signed off on the starting positions for liability, indemnity, insurance.
• The pack survives a procurement read without requiring the agency to "circle back".

Anti-Patterns

• Submitting paragraph-grade pricing language as an "exhibit".
• Inconsistent numbers across exhibits (pricing says 15 % intervention ceiling; SLA says 12 %).
• Open placeholders in the submission ("rate to be agreed").
• Promising a clause in the cover letter that the legal team has not approved.
• Stapling the SaaS MSA on without the agent addendum.
• Drafting credits "in the spirit of partnership" instead of as a formula.
• Forgetting the FX corridor in an Africa-context engagement.

Outputs

• Contract Language Annex (proposal artefact).
• MSA / SoW Insertion Pack (for legal team).
• Internal-consistency checklist signed.
• Legal-team sign-off recorded.

References

• ../references/ai-agent-pricing-exhibit-template.md
• ../references/ai-agent-sla-exhibit-template.md
• ../references/ai-agent-credit-and-refund-clauses.md
• ../references/ai-agent-abandonment-and-refund-policy-template.md
• ../references/ai-agent-msa-addendum-template.md
• ../references/ai-agent-dispute-resolution-and-audit-rights.md
• ../references/ai-agent-outcome-pricing-clauses.md
• ../references/ai-agent-vendor-cost-pass-through.md
• ../references/ai-agent-success-definition-language.md
• ../references/ai-agent-renewal-and-true-up-clauses.md
• ../ai-agent-pricing-and-packaging-proposal/SKILL.md
• ../ai-agent-sla-and-credit-schedule/SKILL.md
• ../ai-agent-commercial-packaging/SKILL.md
• ../ai-agent-msa-and-sla-addendum-templates/SKILL.md
• ../10-financial-proposal/SKILL.md