dcp_ai.crewai — CrewAI Integration¶

DCP integration with CrewAI for multi-agent governance. Each agent has its own DCP passport and audit trail, with support for combining trails across the entire crew.

DCPCrewAgent and DCPCrew keep DCP records next to a crew: hash-chained audit entries in the V2 shape, lifecycle and succession records. They do not sign anything and do not run CrewAI themselves. For signed, verifiable citizenship bundles of a CrewAI crew, use the core SDK as in docs/QUICKSTART_CREWAI.md.

Supported DCP Specifications¶

DCP-01 through DCP-09, partially: unsigned records in the shapes of these specifications. It does not implement them in full.

Installation¶

Published on PyPI (26/10/2026). dcp-ai 2.10.0 includes the implementation described here. Upgrade from 2.9.0 and refresh your environment or lockfile. See Python release notes.

Version 2.10.0 shares the crew session nonce with eligible agents as described below, correcting 2.9.0. These helpers still produce unsigned records and do not run CrewAI; use the SDK signing APIs for verifiable bundles.

pip install "dcp-ai[crewai]>=2.10.1,<3"

For source development:

git clone https://github.com/dcp-ai-protocol/dcp-ai.git
pip install -e "/path/to/dcp-ai/sdks/python[crewai]"

Quickstart¶

from dcp_ai.crewai import DCPCrewAgent, DCPCrew

# Fields of the principal's RPR and of each agent's passport, in their V2 form.
# DCPCrewAgent reads human_id, agent_id and capabilities; it does not need full,
# signed payloads (those also carry session_nonce and keys).
rpr = {
    "dcp_version": "2.0",
    "human_id": "rpr:human-001",
    "legal_name": "Example Operator",
    "entity_type": "natural_person",
    "jurisdiction": "ES",
    "liability_mode": "owner_responsible",
    "override_rights": True,
    "issued_at": "2026-01-01T00:00:00Z",
    "expires_at": None,
}


def passport(agent_id: str, capabilities: list[str], risk_tier: str) -> dict:
    return {
        "dcp_version": "2.0",
        "agent_id": agent_id,
        "principal_binding_reference": rpr["human_id"],
        "capabilities": capabilities,
        "risk_tier": risk_tier,
        "status": "active",
        "created_at": "2026-01-01T00:00:00Z",
    }


researcher = DCPCrewAgent(
    role="researcher",
    passport=passport("agent:researcher-001", ["browse", "api_call"], "low"),
    rpr=rpr,
    goal="Research and gather information",
    backstory="Expert in information retrieval",
)

writer = DCPCrewAgent(
    role="writer",
    passport=passport("agent:writer-001", ["file_write"], "medium"),
    rpr=rpr,
    goal="Produce high-quality content",
    backstory="Professional writer with AI experience",
)

# Create crew with DCP governance
crew = DCPCrew(agents=[researcher, writer], verbose=True)

# Record each agent's participation in the task
result = crew.kickoff(task="Research AI trends and write a summary")
print(result)

# Combined audit trail (ordered by timestamp)
trail = crew.get_combined_audit_trail()
for entry in trail:
    print(f"[{entry['agent_id']}] {entry['intent_id']} -> {entry['outcome']}")

API Reference¶

`DCPCrewAgent`¶

A crew member's DCP record: its passport, its principal's RPR and its own audit trail.

DCPCrewAgent(
    role: str,                     # Agent role in the crew
    passport: dict[str, Any],     # DCP Agent Passport
    rpr: dict[str, Any],          # Responsible Principal Record
    secret_key: str = "",         # Kept on the agent; DCPCrewAgent does not sign with it
    session_nonce: str = "",      # Session nonce; a fresh one when empty, which a DCPCrew replaces with its own
    goal: str = "",               # Agent goal
    backstory: str = "",          # Agent context/backstory
)

Methods¶

Method	Signature	Description
`log_action(action_type, outcome, evidence?, impact?, data_classes?)`	`(str, str, dict?, str?, list?) -> dict`	Records an action as an audit entry in the V2 shape, with hash-chaining (the intent behind it is hashed, not kept). Raises if decommissioned.
`get_audit_trail()`	`() -> list[dict]`	Returns the agent's audit trail
`commission(purpose, capabilities, risk_tier)`	`(str, list?, str?) -> dict`	Commission the agent (DCP-05 §3.1)
`report_vitality(metrics)`	`(dict) -> dict`	Report vitality metrics (DCP-05 §4.1)
`decommission(mode, reason)`	`(str, str) -> dict`	Decommission the agent (DCP-05 §5.1)
`create_testament(successor_preferences, memory_classification)`	`(list, str) -> dict`	Create digital testament (DCP-06 §3.1)

`log_action`¶

entry = researcher.log_action(
    action_type="api_call",
    outcome="success",
    evidence={"url": "https://api.example.com", "status": 200},
)
# entry contains chained intent_hash and prev_hash

Each audit entry includes:

intent_hash: SHA-256 of the associated intent
prev_hash: "GENESIS" for the first entry, SHA-256 of the previous entry for subsequent ones

`DCPCrew`¶

Multi-agent crew with DCP governance.

DCPCrew(
    agents: list[DCPCrewAgent],   # List of DCP agents
    session_nonce: str = "",      # Crew nonce, a fresh one when empty; given to every agent that has not logged and was not given its own
    verbose: bool = False,        # Verbose logging
)

Methods¶

Method	Signature	Description
`kickoff(task)`	`(str) -> dict`	Records one audit entry per agent for the task; it does not run CrewAI
`get_combined_audit_trail()`	`() -> list[dict]`	Combined trail from all agents, ordered by timestamp
`get_agent_bundles()`	`() -> dict[str, list[dict]]`	Individual trails by `agent_id`
`commission_all(purpose?)`	`(str?) -> list[dict]`	Commission all agents in the crew (DCP-05 §3.1)
`succession(from_agent, to_agent)`	`(DCPCrewAgent, DCPCrewAgent) -> dict`	Create succession record (DCP-06 §4.1)
`delegate(human_id, agent, authority_scope)`	`(str, DCPCrewAgent, list) -> dict`	Unsigned delegation record for an agent (DCP-09 §3.1); a signed mandate comes from `create_delegation_mandate` in `dcp_ai.v2`
`get_lifecycle_summary()`	`() -> list[dict]`	Get lifecycle state of each agent

`get_combined_audit_trail()`¶

Combines and chronologically orders the audit trails of all agents. It is a view for reading: each entry is a copy with an added _agent_role field, which is not part of the V2 audit entry, and the combined list is not a hash chain (each agent's entries chain only within its own trail). Do not canonicalize, hash or sign these copies as audit entries.

combined = crew.get_combined_audit_trail()
# [
#   {"agent_id": "agent:researcher-001", "timestamp": "...", "intent_id": "...", "outcome": "...", "_agent_role": "researcher", ...},
#   {"agent_id": "agent:writer-001", "timestamp": "...", "intent_id": "...", "outcome": "...", "_agent_role": "writer", ...},
#   ...
# ]

`get_agent_bundles()`¶

Returns trails separated by agent:

bundles = crew.get_agent_bundles()
# {
#   "agent:researcher-001": [entry1, entry2, ...],
#   "agent:writer-001": [entry3, entry4, ...],
# }

Per-Agent Trails¶

Continuing the quickstart, each agent keeps its own hash-chained trail:

crew = DCPCrew(agents=[researcher, writer])
crew.kickoff(task="Analyze the AI market")

# Unsigned records: seal signed bundles with the core SDK (docs/QUICKSTART_CREWAI.md)
for agent_id, trail in crew.get_agent_bundles().items():
    print(f"\n--- {agent_id} ({len(trail)} entries) ---")
    for entry in trail:
        print(f"  {entry['intent_id']}: {entry['outcome']}")

Development¶

From a checkout:

cd sdks/python
pip install -e ".[crewai,dev]"
pytest -v

Dependencies¶

dcp-ai — DCP SDK (merkle, models, bundle)
crewai — Multi-agent framework (installed by the extra; dcp_ai.crewai itself does not import it)

License¶

Apache-2.0

dcp_ai.crewai — CrewAI Integration¶

Supported DCP Specifications¶

Installation¶

Quickstart¶

API Reference¶

DCPCrewAgent¶

Methods¶

log_action¶

DCPCrew¶

Methods¶

get_combined_audit_trail()¶

get_agent_bundles()¶

Per-Agent Trails¶

Development¶

Dependencies¶

License¶

`DCPCrewAgent`¶

`log_action`¶

`DCPCrew`¶

`get_combined_audit_trail()`¶

`get_agent_bundles()`¶