Install payload
Install payload is broadly covered in current results.
100% (2/2)
Browse the directory
/
Showing 2 resources for "calibration"
Saved
/ search · [ ] view density · esc clear input
Active
Rollout signal scan
2 rollout risk signals in current results
Biggest gaps: metadata review, package integrity. 0 entries have 2+ required gaps.
good
Most at-risk entries in this view
Adoption queue
Browse adoption queue · balanced
0/2 visible results are ready for staged adoption under this preset.
ready 0caution 2hold 0
Inspect AI Benchmark Rubric Agent
1 blockers: Metadata review
caution
50/100
Request metadata review from maintainers or internal owners.
Collect package checksum or signed artifact information.
agents/inspect-ai-benchmark-rubric-agent · trust review · confidence 67%
Phronesis
1 blockers: Metadata review
caution
50/100
Request metadata review from maintainers or internal owners.
Collect package checksum or signed artifact information.
mcp/phronesis-hermes · trust review · confidence 67%
Decision confidence
Decision confidence scan · balanced
0/2 results are high-confidence for the selected preset.
high 0medium 2low 0
Inspect AI Benchmark Rubric Agent
Address Metadata review, Package integrity before broader rollout.
medium
54/100
Missing: Metadata reviewMissing: Package integrity
agents/inspect-ai-benchmark-rubric-agent · trust review
Phronesis
Address Metadata review, Package integrity before broader rollout.
medium
54/100
Missing: Metadata reviewMissing: Package integrity
mcp/phronesis-hermes · trust review
Freshness distribution
Current results are broadly fresh
Median age 46 days; all 2 scanned entries are within 90 days.
Source-backed agent for designing Inspect AI benchmark tasks, datasets, solver plans, scorer rubrics, model matrices, eval logs, and release-quality prompt evaluation decisions.
Phronesisby Sustainable Finance Partners, LLC · submitted by phronesis-cc-bot
Decision-assurance layer for AI agents: returns calibrated Decision Assets and an auditable action boundary (approve / review / block / seek-evidence) plus a verifiable Decision Receipt before high-stakes actions. Advisory only — never transacts or modifies identity. MCP tools: health_check and bench_query.