Skip to content

parlant

AI Agents & Assistants

An open‑source interaction control harness that gives developers fine‑grained, production‑ready governance of customer‑facing AI agents.

Python Latest v3.3.2 · 1mo ago Security brief →

Features

  • Provides precise conversational control and behavior consistency for B2C/B2B interactions
  • Enforces governance, policy constraints, and brand voice at runtime
  • Optimizes context engineering to deliver only the relevant information per turn

Security Response History

1 CVE
CVE Severity Disclosed Patched (this tool) vs Ecosystem Median
CVE-2026-42208 KEV critical
CVSS 9.8
2026-05-08 Unpatched Unpatched

Recent releases

View all 10 releases →
v3.3.2 Mixed
Notable features
  • PARLANT_TOOL_CALL_TIMEOUT environment variable to control plugin tool-call timeout
  • Improved diagnostic error messages from EmcieService
  • Enhanced active journey descriptions in canned response draft prompts
Full changelog

[3.3.2] - 2026-04-28

Added

  • Add PARLANT_TOOL_CALL_TIMEOUT environment variable to control the plugin tool-call timeout

Changed

  • Include guideline description (in addition to the condition) in matcher prompts
  • Render active journeys' descriptions in the canned response draft prompt
  • Improve error messages surfaced by EmcieService to be more diagnostic
  • Relax the aiopenapi3 pin to >=0.8.1,<0.9.0

Fixed

  • Fix redundant batch creation in ToolCaller across guideline matches, which in some agents caused a lot of latency unnecessarily
  • Fix WebSocketLogger event loop starvation which progressively blocked the async event loop, causing increasing latency over time
  • Fix p.Server.current becoming unavailable in the resolve_tool plugin endpoint under Uvicorn 0.39+. Context vars are now restored at the start of resolve_tool, mirroring the existing pattern in call_tool, so choice providers depending on p.Server.current work again
  • Fix MCP tool clients getting stuck after sessions go stale. Clients now reconnect automatically instead of failing subsequent calls

Security

  • Upgrade dependencies to address known CVEs: authlib (>=1.6.11), requests (>=2.33.0), fastmcp (>=3.2.0), litellm (>=1.83.0), pytest (>=9.0.3), pyjwt (>=2.11.1), and constrain transitive deps — aiohttp, cryptography, pillow, pyopenssl, werkzeug, Mako, pyasn1, python-multipart, orjson, Pygments, diskcache
  • Upgrade chat frontend: vite (>=7.3.2) and override transitive deps: picomatch, lodash, flatted, brace-expansion, immutable, yaml
v3.3.1 New feature
Notable features
  • AnyOf/AllOf modifiers for explicit tag dependency semantics
  • depend_on_any() for OR-based dependency relationships
  • Event loop health monitoring on /healthz endpoint with latency tracking
v3.3.0 Breaking risk
Breaking changes
  • Changed tags field type from Sequence[TagId] to Sequence[Tag]
  • Changed reevaluate_after() to accept multiple tools and return Sequence[Relationship]
  • Changed Tag.preamble() to return Tag object instead of TagId
Notable features
  • Per-agent planners
  • Tag-based relationships and dependencies
  • Transient guidelines
v3.2.1 Mixed
⚠ Upgrade required
  • `attach_tool()` deprecated; use `create_guideline()` or `create_observation()` with `tools` parameter instead
Notable features
  • Optional `dependencies` parameter for guideline, observation, and journey creation methods
  • `exclude()` method added as alias for `prioritize_over()` on guidelines and journeys
  • `tools` parameter added to `create_observation` methods
Full changelog

[3.2.1] - 2026-02-17

Added

  • Add optional dependencies parameter to guideline, observation, and journey creation methods
  • Add exclude() as an alias for prioritize_over() on guidelines and journeys
  • Add tools parameter to create_observation methods

Changed

  • Deprecate attach_tool() in favor of create_guideline()/create_observation() with tools parameter

Fixed

  • Preserve draft message language during canned response recomposition
  • Fix server hang when an exception occurs during setup
  • Fix canned response field extraction to handle falsy values
v3.2.0 New feature
Notable features
  • Add labels to Guidelines, Journeys, JourneyNodes, and Sessions with automatic propagation from matched entities
  • Add on_match and on_message hooks to journeys for lifecycle callbacks
  • Add streaming message output mode
Full changelog

[3.2.0] - 2026-02-09

Added

  • Add labels to Guidelines, Journeys, JourneyNodes, and Sessions for categorization and filtering
  • Add automatic session label propagation from matched entities (guidelines, observations, journeys)
  • Add track parameter to guidelines to control "previously applied" tracking
  • Support multiple targets in prioritize_over() and depend_on() methods
  • Add field_dependencies to canned responses for explicit field availability requirements
  • Add attach_retriever() to Guideline, Journey, and JourneyState for conditional data retrieval
  • Add on_match and on_message hooks to journeys for lifecycle callbacks
  • Add per-agent preamble configuration (custom examples and instructions)
  • Add separate default greeting responses for first agent message in fluid mode
  • Add streaming message output mode
  • Allow specifying custom journey node ID
  • Add matched guidelines/journey states to completion ready event

Changed

  • Make condition optional for SDK guidelines
  • Tweak default preamble examples
  • Soften log levels for relational guideline resolver
  • Add activated/skipped logs to custom guideline matcher batches

Fixed

  • Fix websocket warning upon startup
  • Fix agent intention proposer (guidelines were getting rewritten incorrectly)
  • Fix multiple customer guideline matchers not working
  • Fix bug with context variable access in SDK

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

About

Stars
18,094
Forks
1,536
Languages
Python Gherkin TypeScript

Install & Platforms

Install via
pip

Community & Support

Alternative to

Ada Decagon Sierra

Beta — feedback welcome: [email protected]