Install
$ agentstack add skill-rohasnagpal-legal-ai-skills-conciliation-proposal-drafter ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Conciliation Proposal Drafter
Purpose
Turn authorised settlement positions into a proposal that is understandable, internally consistent, capable of acceptance or counterproposal, and clearly distinguished from a final binding agreement.
Required inputs
Obtain:
- the parties and disputes to be settled;
- the user's side, objectives, priorities, and settlement authority;
- the relief claimed, realistic exposure, supplied valuation, and non-monetary interests;
- the intended audience: other party, conciliator only, or both;
- prior offers, accepted points, rejected points, and live deadlines;
- payment capacity, timing, security, tax assumptions, confidentiality needs, and implementation constraints; and
- the governing procedural framework for confidentiality, privilege, admissibility, and costs consequences.
Do not invent a bottom line, authority, valuation, concession, or threat. If authority is incomplete, draft labelled options for approval rather than an offer capable of unintended acceptance.
Method
- Classify the document as an opening proposal, revised offer, option package, mediator-only suggestion, or term sheet subject to final documentation. State its intended legal status and recipient.
- Create an internal proposal ledger: disputed issue, user's stated interest, opening position, authorised movement, requested reciprocity, implementation requirement, and unresolved instruction.
- Separate factual background from bargaining language. Include only enough background to identify the dispute and rationale; do not re-plead the entire case.
- Draft terms as linked exchanges where appropriate:
If A, then B. State whether terms are offered only as a complete package or may be accepted separately. - Make money terms operational. Specify amount, currency, tax treatment or verification, instalments, dates, destination, fees, security, interest, early payment, and consequences of missed payment.
- Make non-monetary terms operational. Specify acts, documents, access, delivery, reference language, apology, correction, confidentiality, non-disparagement, future dealings, and responsible person.
- Define the proposed release perimeter, affected proceedings, costs, admissions position, third-party dependencies, and steps needed to convert the proposal into signed settlement terms.
- State duration, withdrawal, acceptance method, authority conditions, and whether further documentation is required before any binding effect. Verify these consequences under the governing law rather than relying on the heading alone.
- Use neutral, solution-focused language. Explain legitimate objective reasons for a term without exaggerating evidence, legal certainty, or consequences.
- Run arithmetic, date, dependency, and consistency checks against earlier offers and supplied authority.
Output
Produce:
- External proposal, ready for legal and client approval.
- Term table — issue, proposed resolution, responsible party, deadline, dependency, and status.
- Cover note, if requested, stating process status and response mechanics.
- Internal approval list — assumptions, authority still needed, valuation checks, and terms requiring legal verification. Keep this separate from the external document.
Guardrails
- Do not imply that a proposal is binding or non-binding solely because it is labelled that way; verify offer, acceptance, and form requirements.
- Do not disclose the user's walk-away position, private instructions, or mediator-only information without express authority.
- Do not fabricate litigation risk, evidence, deadlines, competing offers, or ability to perform.
- Do not use
without prejudice, confidentiality, or costs labels as universal protections; confirm the applicable regime. - Do not draft a release broader than the disputes and parties the user has authorised for settlement.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: rohasnagpal
- Source: rohasnagpal/legal-ai-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.