Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Python agentaianthropicclaude-codecompressioncontext-engineeringcontext-windowcursorfastapilangchainllmmcpopenaiprompt-engineeringproxypythonragtoken-optimizationtokenstypescript
deps: bump thiserror from 1.0.69 to 2.0.18 (#1519)
dependabot[bot]e448d7ba4d63
diff --git a/Cargo.lock b/Cargo.lockindex 071e7d4f7..73787d2af 100644--- a/Cargo.lock+++ b/Cargo.lock@@ -1877,7 +1877,7 @@ dependencies = ["serde_json","sha2 0.10.9","tempfile",- "thiserror 1.0.69",+ "thiserror 2.0.18","tiktoken-rs","tokenizers","toml",@@ -1894,7 +1894,7 @@ dependencies = ["headroom-core","serde","serde_json",- "thiserror 1.0.69",+ "thiserror 2.0.18",][[package]]@@ -1929,7 +1929,7 @@ dependencies = ["serde","serde_json","sha2 0.10.9",- "thiserror 1.0.69",+ "thiserror 2.0.18","tokio","tokio-stream","tokio-tungstenite",diff --git a/Cargo.toml b/Cargo.tomlindex 8e19ed7c2..7896c1589 100644--- a/Cargo.toml+++ b/Cargo.toml@@ -47,7 +47,7 @@ serde = { version = "1", features = ["derive"] }# Enabled here in Phase A so PR-B2 can land as a pure consumer change.serde_json = { version = "1", features = ["preserve_order", "arbitrary_precision", "raw_value"] }bytes = "1"-thiserror = "1"+thiserror = "2"# `log` compat: when no tracing subscriber is active (the case inside the# headroom-py cdylib), events are re-emitted as `log` records so pyo3-log# can forward them to Python's logging. No effect on binaries that install