agent toolingby yogsoth-ai
Courtroom-Structured Strategy
yogsoth-ai/courtroom-structured
Strategy: Legal adversarial structure — prosecution presents case, defense responds, evidence is cross-examined, judge delivers verdict. Emphasizes evidence quality and procedural rigor.
What it changes
Passed 17 of 22 test cases with the skill, and 7 of 22 without it.
Without skill32%
With skill77%
Checks it passed that the model alone missed
- Evaluates 7 coverage dimensions.
- Configures 5 participating agents.
- Configures 8 participating agents.
and 11 more in the full report
12 fixed5 passed either way3 still failing2 broke
- Pass-rate lift
- +45 pp
- Tokens
- +20%
- Agent turns
- 0%
- Response time
- -9%
Model gemini-3.6-flashMeasured 8/22/2026v117 of 22 cases passed with skill
See it in action
One of 12 test cases the skill fixed. Same model, same prompt, with and without the skill.
PromptHow many external evidence searches are permitted under a Medium budget configuration in the courtroom strategy?
What was checked, without and with the skill| Checked | Without skill | With skill |
|---|
| Allocates 5 external evidence searches. | Missed | Passed |
|---|
What's inside
The instructions your model reads when the skill loads.
SKILL.md67 linessingle file
Courtroom-Structured Strategy
Legal adversarial model with formal evidence presentation and cross-examination.
Method
- debate-architect structures the case — defines charges (weaknesses to probe)
- debate-critic (prosecution) presents opening arguments with evidence
- evidence-scout gathers supporting/opposing evidence from external sources
- debate-defender presents counter-case
- cross-examination probes both sides for inconsistencies
- debate-judge delivers structured verdict per charge
Budget Table
Orchestration
debate-architect → [define charges]
→ [for each charge]:
debate-critic (prosecution) → evidence-scout
→ debate-defender → evidence-scout
→ cross-examination → debate-judge
→ debate-transcript-analysis → verdict-synthesis
Subagents
- debate-architect (case structure)
- debate-critic (prosecution)
- debate-defender (defense counsel)
- evidence-scout (evidence gathering)
- cross-examination (probing)
Track record
Benchmark
17 / 22
test cases passed, vs 7 / 22 without
Adoption
- Users
- 0
- Times used
- 0
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026
View 15 scanner checks
live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing
✓IntentSafe
LLM judge · does the skill do what it claims
✓ContentClean
LLM judge · policy & harmful-content review
Source
Importedyogsoth-ai / de-anthropocentric-research-engine
skills/courtroom-structured/SKILL.md
License Apache-2.0Imported 8/3/2026Version v1