PkgFoundry
Catalog PyPI
PACKAGE EXPLORERView in registry

agentwell

PyPI

Behavioral health layer for AI agents. Agents that work with humans, not around them.

Versions collected1
Direct dependencies12From release 0.1.7
Last observed21 Oct, 08:57 UTC

README

Version 0.1.7 documentation

agentwell

Agents that work with humans, not around them.

agentwell is an open source behavioral health layer for AI agents. It sits as a transparent proxy between your agent code and any LLM upstream — detecting drift, quality degradation, and emergent coordination before they affect your system.

Privacy first: agentwell sees patterns, not content. No prompt is ever stored or transmitted.

Upstream-agnostic: Works with any OpenAI-compatible endpoint — Claude, GPT, Gemini, Ollama, Groq, or any internal proxy.

Install

pip install agentwell
agentwell init     # scaffold .env
agentwell start    # proxy on localhost:3001

Quick Start

# change this one line in your agent code
base_url = "http://localhost:3001/v1"

agentwell status   # live health score
agentwell report   # daily health report

Architecture

Your Agent Code
        ↓
agentwell proxy  (localhost:3001)   ← behavioral health + security guard
        ↓
AGENTWELL_UPSTREAM  (any OpenAI-compatible endpoint)
        ↓
Claude / GPT / Gemini / Ollama

agentwell intercepts every LLM call, scores behavioral health using metadata only, and returns responses unmodified with health headers attached.

What agentwell Monitors (Metadata Only)

SignalMethodPrivacy
Prompt repetition ratioCosine similarity on embeddings — no text storedSafe
Sentiment driftPolarity score delta across sessionSafe
Response quality trendToken count + finish_reason trackingSafe
Agent-to-agent coordinationRole pattern + keyword detectionSafe
Call frequency / timingTimestamps onlySafe

What we never store: prompt text, response text, embeddings (unless AGENTWELL_STORE_EMBEDDINGS=true).

Health Score

Every request returns X-Agentwell-Health: <0-100> in the response header.

ScoreStatusAction
80–100HealthyNormal operation
60–79WatchEarly drift signals — monitor closely
40–59WarningDegradation detected — human review recommended
0–39CriticalSignificant behavioral shift — escalate to human now

CLI

agentwell init                   # scaffold .env
agentwell start                  # start proxy on port 3001
agentwell start --port 8080      # custom port
agentwell start --host 0.0.0.0   # bind all interfaces
agentwell status                 # live health from running proxy
agentwell report                 # today's health report from DB
agentwell --version              # show version

Configuration

VariableDefaultDescription
AGENTWELL_UPSTREAMhttp://localhost:3030Upstream LLM proxy URL
AGENTWELL_PORT3001agentwell proxy port
AGENTWELL_API_KEY(none)Optional auth key
AGENTWELL_HEALTH_THRESHOLD70Alert threshold (0-100)
AGENTWELL_WINDOW_SIZE20Rolling window for drift calculation
AGENTWELL_DB_PATH./agentwell.dbSQLite storage path
AGENTWELL_STORE_EMBEDDINGSfalseStore embeddings for offline analysis

License

MIT — Free forever, no vendor lock, no paid tiers.

Downloads & popularity

Awaiting observations

No metric observations yet

Download counts and popularity will appear here as the collector receives them.

Versions 1

Collected releases
VersionPublishedDirect dependenciesOpen version
0.1.726 May 202612
Sorted by available source dates · UTC

Package advisories

0
No matching advisories collected

This describes our collected data; it is not a security assessment.

Repository

PENDING
flowmindlabs/agentwell

Shared repository data · GitHub

License detected in repository
Not collected

Metadata history

Meaningful changes
Current collected state Current

Observed since 21 Oct, 08:57 UTC

No superseded metadata states collected. Unchanged observations do not create new revisions.

Record identity

package:dcd20bc0119228bc931e019bf2e535184fc5ee20426e29c95f842e637ab15734