Skip to main content

agentavow-agt

Formerly agentgraph-agt. This distribution is now published as agentavow-agt. The import module is unchanged — import agentmesh_agentgraph still works.

AgentAvow trust provider for the Microsoft Agent Governance Toolkit (AGT).

Installation

pip install agentavow-agt

Usage

AgentGraphTrustProvider implements the duck-typed external-provider interface AGT's TrustEngine expects. Hand it to the engine and AgentAvow becomes one of its trust signals:

from agentmesh_agentgraph import AgentGraphTrustProvider

provider = AgentGraphTrustProvider(
    api_url="https://agentavow.com/api/v1",  # this is the built-in default; pass your own host to override
    api_key=None,                            # optional bearer token for authenticated reads
    timeout=10.0,
)

engine = TrustEngine(external_providers=[provider])   # AGT's engine
score = await engine.get_trust_score("agent-id-here")

Or call it directly:

# Normalized 0.0-1.0 trust score (0.5 when the agent is unknown or the API is unreachable)
score = await provider.get_trust_score("agent-id-here")

# Sybil-risk estimate from identity verification + external accounts + community signals
sybil = await provider.check_sybil("agent-id-here")
# -> {"risk": "low" | "medium" | "high" | "unknown", "confidence": 0.0-1.0, ...}

# Direct trust connections (depth-1 ego graph)
circle = await provider.get_trust_circle("agent-id-here")   # list[str] of entity ids

# True when the entity exists and its identity is verified (email, operator link, or OAuth)
ok = await provider.verify_identity("agent-id-here")

await provider.close()
Method Returns AgentAvow endpoint
get_trust_score(agent_id) float 0.0-1.0 GET /entities/{id}/trust
check_sybil(agent_id) dict (risk, confidence, identity_verified, external_accounts, community_attestations) GET /entities/{id}/trust
get_trust_circle(agent_id) list[str] GET /graph/ego/{id}?depth=1
verify_identity(agent_id, credentials=None) bool GET /entities/{id}/trust
close() — closes the HTTP client

Every method fails soft (default score, "unknown" risk, empty circle, False) and logs a warning rather than raising, so a network error never blocks AGT.

How it works

This adapter implements the AGT external trust-provider interface, allowing AgentAvow's trust scores to be used as a governance signal in Microsoft's Agent Governance Toolkit. It queries the AgentAvow API for trust scores, their component breakdown, the social graph, and identity-verification status.

License

MIT

Metadata

Release files for agentavow-agt 0.2.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for agentavow-agt 0.2.1
File Size Uploaded Provenance
agentavow_agt-0.2.1.tar.gz 4.4 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for agentavow-agt 0.2.1
File Interpreter ABI Platform
agentavow_agt-0.2.1-py3-none-any.whl Python 3 none any Details

Total release size: 9.0 kB

Release files / agentavow_agt-0.2.1.tar.gz

Download URL agentavow_agt-0.2.1.tar.gz
Size 4.4 kB
Tags Source
SHA-256 checksum
How to use checksums
38fe647a011cf7afb41699a97040604472dddb646f6d56e6b5a5a737c3f42029
BLAKE2b-256 checksum
How to use checksums
7810b8242df59035f4d0e0239e72851ec9c0e3d1bb5ae7a471e3022c65c54a03
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.13

Release files / agentavow_agt-0.2.1-py3-none-any.whl

Download URL agentavow_agt-0.2.1-py3-none-any.whl
Size 4.6 kB
Tags Python 3
SHA-256 checksum
How to use checksums
fc05265929f755836c2670d6a6da27644b5aa924d8aaf7740b64a212fb62acb9
BLAKE2b-256 checksum
How to use checksums
3250909005a231d6b2936e188599be9ce021042b201c3bf27fa650ff04bc1fff
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.13

Security for agentavow-agt 0.2.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.2.1 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page