hypothesise
No liveness check has reached it yet; it is registered on GitHub and PyPI and has been in the index since 11 Oct 2026. How this is checked
Hypothesise is an open-source behavioral security layer for AI agents. It observes tool calls, learns normal agent behavior, and flags potentially dangerous deviations for developers and security teams.
Inferred · not functionally tested
Overview
6 featuresPurpose: Detecting anomalous or dangerous tool calls made by AI agents.
Inferred · not functionally tested
Audience: AI security engineers and developers building AI agents
Inferred · not functionally tested
Functions: monitoring, agents
Inferred · not functionally tested
Interfaces: API: indicated (inferred, not tested) · MCP: indicated (inferred, not tested) · CLI: indicated (inferred, not tested) · Self-hosting: indicated (inferred, not tested)
Recorded constraints: pricing: open_source · license: Apache-2.0 · platforms: CLI · deployment: cli, self_hosted, api_only
Constraint provenance is unknown; confirm requirements with the publisher.
Record sources: pypi.org · github.com. These links do not verify the individual claims.
hypothesise is an AI & LLM security project. Inferred · not functionally tested: It focuses on detecting anomalous or dangerous tool calls made by AI agents. Inferred · not functionally tested: hypothesise is an open-source project aimed at AI security engineers and developers building AI agents. Basis unknown · not verified: hypothesise is open source under the Apache-2.0 license. Basis unknown · not verified: It ships for the command line and API, and it can be self-hosted.
Behind hypothesise is Siddhant Shah, and it first shipped in 2026. Development happens publicly on GitHub with 22 commits in the last 90 days. Inferred · not functionally tested: Among its 6 catalogued features are tool call monitoring, behavior learning, and anomaly detection. Inferred · not functionally tested: Catalogued interfaces include an MCP server.
Summary written by a language model from the project’s public pages.
Tasks: Inferred · not functionally tested
- Tool call monitoring
- Behavior learning
- Anomaly detection
- Dangerous deviation alerts
- AI agent observability
- MCP support
Topics: Inferred · not functionally tested
Built with & integrations
- Claude Code
- CLAUDE.md
Trust & compliance
Indexing history
1What PulseGate has recorded for this listing
- Indexed11 Oct · 05:27 UTChypothesise seen via PyPI Bulk EnumeratorSource: PyPI Bulk Enumerator · Open
Frequently asked questions about hypothesise
- What is hypothesise?
- Inferred · not functionally tested: Hypothesise focuses on detecting anomalous or dangerous tool calls made by AI agents. It is catalogued under AI & LLM security on PulseGate.
- Who is hypothesise for?
- Inferred · not functionally tested: hypothesise is an open-source project built for AI security engineers and developers building AI agents.
- Is hypothesise free?
- Basis unknown · not verified: Yes — hypothesise is open source under the Apache-2.0 license and free to use.
- What platforms does hypothesise run on?
- Basis unknown · not verified: hypothesise runs on the command line and API. It can also be self-hosted.
- Is hypothesise still maintained?
- The GitHub repository shows 22 commits in the last 90 days.
- What are alternatives to hypothesise?
- Similar projects tracked by PulseGate include mcp-runtime-audit, spillage, and sentinelforge-ai.mcp-runtime-auditspillagesentinelforge-ai
- Who makes hypothesise?
- hypothesise is developed by Siddhant Shah.
- How long has hypothesise been around?
- hypothesise first shipped in 2026.
Also in AI & LLM security
Same category — not a similarity match