News

What's new at ReclaimLLM.

Announcements, feature releases, and technical updates — written for the developers and engineers who use RCLM every day.

feature

Sep 21, 2026

Layered tool-result compression with exact local recall

ReclaimLLM now reduces repeated and verbose tool output with provider-aware transforms, bounded state, and content-addressed local recall handles.

Read →

feature

Sep 21, 2026

Capture LLM API Calls and Coding-Agent Sessions in One Model

ReclaimLLM now supports OpenTelemetry API workloads alongside native capture for six coding agents, with shared search, analytics, and governance.

Read →

feature

Sep 16, 2026

Capture Existing LLM API Calls with OpenTelemetry

Export provider-neutral LLM traces into ReclaimLLM without proxying inference traffic or sharing provider credentials.

Read →

feature

Sep 1, 2026

Deferred User Account Deletion with a 30-Day Grace Period

Self-service account deletion for Free and Paid tiers with immediate session hiding, halted renewals, and one-click restoration.

Read →

feature

Aug 27, 2026

Proxy operation anomaly detection for enterprise gateway traffic

Organization admins can review unusual proxy-operation volume and first-seen operations with linked session evidence.

Read →

feature

Aug 18, 2026

Compare Different Models on Your Real LLM Workloads

Replay captured proxy sessions against enabled models, measure classification agreement, and inspect each result before changing providers.

Read →

feature

Aug 6, 2026

RCLM Bench: Local Comparisons for Coding Models and Compression Tools

RCLM Bench is an open-source CLI and TUI for isolated coding-model runs and offline compression replay on Claude Code and Codex sessions.

Read →

feature

Aug 4, 2026

Verify RCLM's Token Savings on Your Own Sessions with Replay

Four new read-only MCP tools replay RCLM's shipped compression mechanisms over your own captured sessions and report the real token reduction — no model calls, no re-execution, no dashboard required.

Read →

feature

Jul 30, 2026

RCLM Signals: Patterns for Finding AI Workflow Waste

RCLM Signals turns captured sessions into concrete patterns for admins and developers to investigate: repeated restarts, over-exploration, idle gaps, session bloat, policy drift, duplicated work, and model mismatch.

Read →

feature

Jul 28, 2026

Move Full Captured Sessions Between Claude Code and Codex

Transfer the complete ReclaimLLM-captured history into a new Claude Code or Codex session without reducing it to a summary.

Read →

feature

Jul 26, 2026

Downscaling Oversized Images Before They Reach the Model

RCLM's hook layer now detects images in tool results and downscales them before they enter your agent's context — measured token savings on Claude Code, measurement-only on Codex.

Read →

feature

Jul 24, 2026

Range-aware read caching and brevity controls

RCLM now avoids unchanged file rereads and offers an optional Claude Code instruction for shorter responses.

Read →

feature

Jul 22, 2026

Stripe-hosted billing for personal and enterprise plans

Personal users and organization administrators can now review invoices in RCLM while managing payments securely through Stripe.

Read →

feature

Jul 21, 2026

Claude Code statusline: context, rate limits, and peak hours at a glance

rclm-hooks-install now ships a live Claude Code statusline by default, and along the way we fixed a bug that silently broke rclm-hooks-uninstall for every provider.

Read →

feature

Jul 21, 2026

Kimi and DeepSeek on the LLM Gateway, Plus a Way to Try Them

Kimi (Moonshot) and DeepSeek are now supported Gateway providers, and a new Compare Model action lets you test them against a real captured session before switching.

Read →

feature

Jul 15, 2026

Reducing token cost in agentic coding sessions

New opt-in hooks cut redundant tokens in captured sessions, cache-token measurement is now accurate, and a shadow mode lets you see estimated savings before anything is enforced.

Read →

feature

Jun 17, 2026

Encrypted Session Storage for Paid and Enterprise Users

Paid users and enterprise organizations can protect raw session details with encrypted storage, one-time recovery keys, and controlled decrypt access.

Read →

feature

Jun 10, 2026

ReclaimLLM Agent Plugin Marketplaces

ReclaimLLM now ships local plugin marketplace entries for Codex, Claude, and Cursor, backed by the existing rclm MCP server.

Read →

feature

May 16, 2026

Enterprise LLM Gateway

Enterprise admins can now issue team-scoped LLM gateway keys, restrict provider access, and log proxy calls under the right org and team.

Read →

feature

May 10, 2026

Normalized session tool calls

ReclaimLLM now records one normalized row per tool call so stats pages can show tool usage, token attribution, failures, and command-family breakdowns across providers.

Read →

feature

May 7, 2026

Unified Recall and Capture: Introducing the ReclaimLLM MCP Server

Search and retrieve prior sessions directly from your AI coding tools with a single installation flag.

Read →

feature

May 6, 2026

Cursor capture now uses transcripts as the source of truth

RCLM now captures Cursor sessions by reading the Cursor transcript on completion, while using hook events for file edits, lifecycle data, and fallback coverage.

Read →

feature

May 6, 2026

Session Category Classification

CLI sessions now carry a maintenance category so users can filter and review fix, feature, and performance work.

Read →

feature

Apr 28, 2026

Extract Project Names From CLI Sessions

CLI sessions can now be grouped and filtered by a derived project name based on changed file paths.

Read →

feature

Apr 27, 2026

Local Redaction of Sensitive Information

RCLM hooks can now apply cached substitution rules, local-only redactions, and folder exclusions before session uploads leave the machine.

Read →

feature

Apr 27, 2026

OpenClaw session capture via plugin hooks

RCLM now captures OpenClaw sessions through a lightweight plugin that forwards session, LLM, and tool events into the existing hook ingestion pipeline.

Read →

feature

Apr 21, 2026

Introducing Secure Session Sharing via Opaque Tokens

Share your LLM sessions with clients and collaborators effortlessly—no account required, with full control over revocation and expiration.

Read →

feature

Apr 15, 2026

Seamlessly Continue Your Work Across AI Tools

Move your sessions between Claude, Gemini, and other AI tools with our new Session Context Export feature—no more losing progress when you switch environments.

Read →

update

Mar 26, 2026

Find Anything with New Hybrid Semantic Search

We've upgraded session search to understand your intent—now you can find past work using natural language queries, even if you don't remember the exact keywords.

Read →

update

Mar 25, 2026

Richer, More Structured Session Summaries for Everyone

We've upgraded our session summarization engine to provide detailed "What Happened," "Highlights," and "Improvements" for every session, regardless of your tier.

Read →

feature

Mar 23, 2026

Data Residency: Regional Storage and Migration

ReclaimLLM now supports regional data residency, allowing users to store their session data in the EU to meet compliance and legal requirements.

Read →

announcement

Mar 21, 2026

Introducing Our New Three-Tier User System

We've launching Free, Paid, and Enterprise tiers to provide more tailored features and support as you scale your AI development workflow.

Read →

update

Mar 19, 2026

Session Export Delivery via S3 Presigned URL

We've upgraded our session export process to handle larger datasets by moving from direct email attachments to secure, time-limited S3 download links.

Read →

update

Mar 18, 2026

Reliable Session Statistics via Background Task

We've introduced a server-side background task to compute consistent token counts, tool usage, and code impact metrics for every session across all capture sources.

Read →

feature

Mar 15, 2026

Sessions Now Auto-Summarize After Ingest

New sessions are automatically titled, described, and tagged in the background — no manual trigger required.

Read →

update

Mar 15, 2026

Context Compression and Session Analytics

Stay within your LLM's context window and gain deeper insights into your sessions with our new hook-level compression and usage analytics.

Read →