Issues

• [P0] Apply priority and area labels to Milestone 7+ backlog

  askmy-stack/tool-semantics#118 · Filter by author cursor opened on Sep 16, 2026

• [P2] Optional embedding similarity layer for semantic comparisons

  askmy-stack/tool-semantics#117 · Filter by author cursor opened on Sep 16, 2026

• [P3] Confidence intervals for behavioral metrics

  askmy-stack/tool-semantics#115 · Filter by author cursor opened on Sep 16, 2026

• [P3] Dev / test / verified split for benchmarks and heuristics

  askmy-stack/tool-semantics#116 · Filter by author cursor opened on Sep 16, 2026

• [P2] Efficiency regression metrics (calls, retries, latency, tokens)

  askmy-stack/tool-semantics#114 · Filter by author cursor opened on Sep 16, 2026

• [P2] Cross-application / multi-namespace workflow benchmarks

  askmy-stack/tool-semantics#112 · Filter by author cursor opened on Sep 16, 2026

• [P2] Integrity monitoring for illegitimate task success

  askmy-stack/tool-semantics#113 · Filter by author cursor opened on Sep 16, 2026

• [P2] Format-sensitivity fuzz testing for equivalent schemas

  askmy-stack/tool-semantics#111 · Filter by author cursor opened on Sep 16, 2026

• [P2] Behavioral Compatibility Horizon metric

  askmy-stack/tool-semantics#110 · Filter by author cursor opened on Sep 16, 2026

• [P2] Task difficulty and messiness metadata for probes/benchmarks

  askmy-stack/tool-semantics#109 · Filter by author cursor opened on Sep 16, 2026

• [P1] No-tool correctness probes (when agents must not call tools)

  askmy-stack/tool-semantics#107 · Filter by author cursor opened on Sep 16, 2026

• [P1] Final-state verification independent of tool trajectory

  askmy-stack/tool-semantics#105 · Filter by author cursor opened on Sep 16, 2026

• Add demo MCP server variants for safe, breaking, and confusing changes

  askmy-stack/tool-semantics#102 · Filter by author cursor opened on Sep 16, 2026

• Restructure docs and README for capture+eval positioning

  askmy-stack/tool-semantics#101 · Filter by author cursor opened on Sep 16, 2026

• Optional SARIF and HTML report outputs for eval

  askmy-stack/tool-semantics#99 · Filter by author cursor opened on Sep 16, 2026

• Parallel probes, result caching, cost tracking, and scale benchmarks

  askmy-stack/tool-semantics#100 · Filter by author cursor opened on Sep 16, 2026

• Generate human-reviewable probe drafts from a snapshot

  askmy-stack/tool-semantics#98 · Filter by author cursor opened on Sep 16, 2026

• Add lint and audit commands for tool/description quality

  askmy-stack/tool-semantics#97 · Filter by author cursor opened on Sep 16, 2026

• Add tool-semantics init and doctor commands

  askmy-stack/tool-semantics#96 · Filter by author cursor opened on Sep 16, 2026

• Multi-model matrix, temperature sweeps, and reproducibility metadata

  askmy-stack/tool-semantics#94 · Filter by author cursor opened on Sep 16, 2026

• Tool-description sensitivity and catalog-scaling research harness

  askmy-stack/tool-semantics#95 · Filter by author cursor opened on Sep 16, 2026

• Build multi-domain behavioral compatibility benchmark corpus

  askmy-stack/tool-semantics#92 · Filter by author cursor opened on Sep 16, 2026

• Add mutation generator and precision/recall detection metrics

  askmy-stack/tool-semantics#93 · Filter by author cursor opened on Sep 16, 2026

• Capture and diff MCP extensions (and task-capability stubs)

  askmy-stack/tool-semantics#91 · Filter by author cursor opened on Sep 16, 2026

• Structural compatibility diffs for prompts and resources

  askmy-stack/tool-semantics#90 · Filter by author cursor opened on Sep 16, 2026