ATIF-native analytics over Claude Code agent trajectories: convert sessions to ATIF, materialize a corpus, query it with DuckDB.
-
Updated
Oct 10, 2026 - Python
#
agent-trajectories
Here are 12 public repositories matching this topic...
[EMNLP 2026] Messier is a cross-benchmark corpus of AI agent evaluations, linking tasks and results with models, scaffolds, verifiers, scoring rules, and execution trajectories.
benchmark datasets ai-agents item-response-theory psychometrics llm-evaluation agent-evaluation difficulty-prediction agent-benchmarks agent-trajectories
-
Updated
Oct 6, 2026 - Python
Multilingual software-engineering benchmark with pinned Docker environments & reproducible agent trajectories · 多语言软件工程评测基准
multilingual docker benchmark dataset software-engineering code-repair llm-evaluation coding-agent swe-bench agent-trajectories
-
Updated
Aug 18, 2026 - Shell
Copy inflation in multi-turn search agents: 78-92% of generated tokens are copied from retrieved documents and carry inflated log-probabilities, breaking confidence-based methods. Diagnostic toolkit + Retrieval-Grounded Voting. Findings of EMNLP 2026.
calibration confidence llm retrieval-augmented-generation test-time-scaling emnlp2026 search-agents agent-trajectories
-
Updated
Aug 28, 2026 - Python
VCR cassettes for agent trajectories: record agent runs as DAGs, replay the canonical path, only call the LLM for net-new paths
python caching memoization openai replay dag vcr ai-agents llm langchain llmops llm-agents agent-trajectories
-
Updated
Aug 12, 2026 - Python
Local-first trajectory learning for coding agents. Every run makes the next one better.
python cli pi codex ai-agents local-first agent-learning agentic-ai claude-code agent-observability coding-agent developer-tools-ai-agent trajectory-learning agent-trajectories
-
Updated
Oct 2, 2026 - Python
Open rubric and a small, hand-built gold set for judging multi-step web-research agent trajectories. The first dataset from Assayo.
evaluation dataset agents rubric tool-use llm-as-a-judge web-research deep-research agent-trajectories
-
Updated
Jun 8, 2026 - Python
Open rubric and a small, hand-built gold set for judging multi-step web-research agent trajectories. The first dataset from Assayo.
evaluation dataset agents rubric tool-use llm-as-a-judge web-research deep-research agent-trajectories
-
Updated
Jun 8, 2026 - Python
Export Claude Code, Codex and Cursor sessions as ATIF trajectories. Rust core, Python CLI, installable with uv.
-
Updated
Sep 22, 2026 - Rust
Discovering named behaviours in LLM agent trajectories and testing if they flag failing runs early.
hidden-markov-model failure-analysis llm-agents agent-evaluation agent-observability terminal-bench agent-trajectories
-
Updated
Oct 4, 2026 - Python
Independent research on trajectory-aware AI-agent evaluation, delegated authority, control integrity, and failure-preserving reproducibility.
reproducible-research ai-safety human-in-the-loop ai-agents tool-use ai-governance llm-evaluation agentic-ai agent-evaluation control-integrity delegated-authority agent-trajectories
-
Updated
Oct 1, 2026 - Python
Capture coding-agent sessions (Claude Code / Codex) at the source — trajectory + verifiable git environment — and score each session's training value (grounded × rich × focused). Local, deterministic, no model at runtime.
git machine-learning reinforcement-learning provenance dataset developer-tools codex sft coding-agents claude-code llm-training-data agent-trajectories
-
Updated
Aug 25, 2026 - Python
Add this topic to your repo
To associate your repository with the agent-trajectories topic, visit your repo's landing page and select "manage topics."