Install
$ agentstack add skill-rohasnagpal-legal-ai-skills-defence-strategy-planner ✓ 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
Defence Strategy Planner
Build a defensible strategy from the governing law and available record. Separate established facts, client instructions, allegations, reasonable inferences, and unknowns throughout.
Required inputs
- Jurisdiction, charges and alleged conduct
- Applicable substantive, procedural and evidence law, if known
- Complaint, police or prosecution record, orders and disclosed evidence
- Client instructions, including disputed facts and objectives
- Custody, bail or other restrictions and every live deadline
- Prior applications, admissions, offers, counsel advice and case actions
If a critical input is unavailable, identify it and explain how the gap affects the strategy. Do not silently invent facts or law.
Method
- Frame the case. Identify the controlling law, forum, procedural stage, burdens and standards of proof. Verify current law where the answer may have changed.
- Map the prosecution case. For each count, list every element, the evidence said to prove it, admissibility or reliability issues, defence material, and the present assessment. Treat absence of disclosure as unknown rather than proof of absence.
- Build the chronology. Create a source-linked timeline. Mark conflicts in time, place, identity, conduct and subsequent behaviour without overstating what each conflict proves.
- Test competing theories. Develop only theories consistent with the known record and client instructions. For each, state supporting facts, contrary facts, evidence still needed and consequences if rejected. Do not force the evidence into a preferred narrative.
- Plan evidence work. Identify lawful preservation, disclosure, inspection, records, digital evidence, expert, forensic and witness steps. Record provenance and chain-of-custody concerns.
- Plan procedure. Prioritise jurisdiction, charge, disclosure, exclusion, severance, expert, witness and other motions available in the jurisdiction. Give prerequisites, deadlines, evidence required and downside risk.
- Plan trial themes. Set concise themes and issue-based examination topics tied to admissible evidence. Identify impeachment material and likely rehabilitation. Never script false evidence or coach a witness toward a version.
- Compare resolutions. Where lawful and instructed, compare trial, plea, diversion, restorative or settlement-related routes. Address admissions, sentence exposure, collateral consequences, enforceability and voluntariness.
- Create a decision tree. Rank immediate actions, decision points and fallback positions by urgency, value and risk. Assign owners and dates.
Output
Produce:
- A one-page case theory and objectives
- An element-and-evidence matrix for each count
- A verified chronology with conflicts and gaps
- A motion, disclosure and investigation plan
- A witness, expert and trial-theme plan
- A resolution-options matrix with risks and consequences
- A dated action list, decision tree and client questions
Guardrails
- Do not assist with fabrication, a false alibi, coached testimony, destruction or concealment of evidence, witness intimidation or unauthorised contact, evasion, or obstruction.
- Preserve adverse material and explain disclosure or preservation duties.
- Do not treat a contradiction, procedural defect or weak item as conclusive innocence or automatic dismissal.
- Protect privileged strategy, confidential data and protected identities. Use neutral labels in shareable drafts.
- Flag conflicts of interest, capacity concerns and decisions requiring qualified local counsel.
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.