crewAI · Health snapshot

crewAI repository health snapshot

Created · run_mwileer43a2f462a1cc4425e · github.com/crewAIInc/crewAI @ main · commit 364b87959014 · model claude-opus-5-5

A five-minute read on how healthy this repository is: seven ratings, the weakest point under each, and the findings a quick read turned up.

985source files
541test files
35module groups
19CI workflows

1. Summary

crewAI is a Python framework for building multi-agent AI systems: agents with roles and tools are organised into crews and flows that call LLM providers, use memory and knowledge stores, and can be exposed to other agents over A2A or MCP. The repository is a uv workspace of five packages (the core crewai library, crewai-tools, the crewai CLI, crewai-core and crewai-files) plus release tooling; it ships as libraries and a CLI, not as a deployed service.

Overall health: Adequate. Tooling, tests and security controls are well above average for an open-source framework, but every agent run unpickles a file from the working directory that agent tools can write to, and error handling swallows failures in 126 places.

RatingScaleScore
Engineering standards7/10 Good
Security6/10 Fair
Configuration and secrets6/10 Fair
Data and persistence6/10 Fair
Error handling and resilience6/10 Fair
Availability and recovery5/10 Fair
Performance and scale6/10 Fair
Indicative ratings from a quick read. The full assessment works through every criterion in depth and often finds more, so its ratings are usually the same or lower.

2. Architecture at a glance

Dependencies run from tests into the five packages. The core library and crewai-core, the CLI and the core library, and crewai-tools and the core library each import one another in both directions, so the package boundaries are not one-way.

module dependencies 73 77 98 118 147 160 173 178 191 213 225 239 274 334 382 498 499 631 664 2049 no edge drawn lib/cli 1 file lib/cli/src/crewai_cli 97 files lib/cli/tests 55 files lib/cli/tests/authentication 9 files lib/crewai 1 file lib/crewai-core 1 file lib/crewai-core/src/crewai_core 30 files lib/crewai-core/tests 12 files lib/crewai-files 1 file lib/crewai-files/src/crewai_files 36 files lib/crewai-files/tests 15 files lib/crewai-tools 1 file lib/crewai-tools/src/crewai_tools 268 files lib/crewai-tools/tests 10 files lib/crewai-tools/tests/rag 22 files lib/crewai-tools/tests/tools 52 files lib/crewai/src/crewai 552 files lib/crewai/tests 89 files lib/crewai/tests/agents 25 files lib/crewai/tests/cli 23 files lib/crewai/tests/events 9 files lib/crewai/tests/experimental 11 files lib/crewai/tests/hooks 12 files lib/crewai/tests/llms 43 files lib/crewai/tests/mcp 13 files lib/crewai/tests/memory 14 files lib/crewai/tests/rag 20 files lib/crewai/tests/skills 12 files lib/crewai/tests/telemetry 27 files lib/crewai/tests/tools 13 files lib/crewai/tests/tracing 8 files lib/crewai/tests/utilities 41 files lib/devtools 14 files scripts 4 files
Figure 1 · module dependenciesGenerated from modules.cjs, so it matches the tables in this section exactly. Columns are dependency depth: a module sits one column right of everything that calls into it. Line thickness is the number of calls and imports crossing that boundary. The 10 modules under “no edge drawn” take part in none of the edges shown, heaviest first. The accented box is the module with the most files.
ModuleFilesWhat it owns
lib/crewai/src/crewai552The framework: agents, crews, tasks, flows, LLM providers, memory, knowledge, events, telemetry, A2A and MCP.
lib/crewai-tools/src/crewai_tools268Tool integrations (search, scraping, browsers, databases, vector stores) and RAG loaders.
lib/cli/src/crewai_cli97The crewai command: scaffolding, run, train, deploy, login, checkpoint and memory TUIs.
lib/crewai/tests89Core tests at the package root: crew, flow, project loading, checkpoints, multimodal.
lib/cli/tests55CLI tests.
lib/crewai-tools/tests/tools52Tool tests, mostly against recorded HTTP cassettes.
lib/crewai/tests/llms43LLM provider tests.
lib/crewai/tests/utilities41Tests for shared utilities.
lib/crewai-files/src/crewai_files36File inputs for multimodal calls: provider uploaders, upload cache and cleanup.
lib/crewai-core/src/crewai_core30Shared core: auth providers, token manager, platform API client, paths and constants.
lib/crewai/tests/telemetry27Telemetry and span tests.
lib/crewai/tests/agents25Agent and agent executor tests.
lib/crewai/tests/cli23An older copy of the CLI tests, still run against crewai-core.
lib/crewai-tools/tests/rag22RAG loader tests.
lib/crewai/tests/rag20Core RAG client tests.
lib/crewai-files/tests15File handling and uploader tests.
lib/crewai/tests/memory14Memory and storage backend contract tests.
lib/devtools14Release tooling: version bump, tag, release and docs checks.
lib/crewai/tests/tools13Core tool layer tests.
lib/crewai/tests/mcp13MCP client and transport tests.
lib/crewai/tests/hooks12LLM and tool hook tests.
lib/crewai/tests/skills12Skills loading and cache tests.
lib/crewai-core/tests12Shared core tests.
lib/crewai/tests/experimental11Experimental evaluation feature tests.
lib/crewai-tools/tests10Tool package tests outside the tool and RAG folders.
lib/crewai/tests/events9Event bus tests.
lib/cli/tests/authentication9CLI login and identity provider tests.
lib/crewai/tests/tracing8Execution tracing tests.
scripts4Maintenance scripts: docs version freezing, model catalog update, evaluation runners.
(repo root)1Shared pytest fixtures and cassette scrubbing (conftest.py).
lib/crewai-files1Package manifest.
lib/crewai1Package manifest.
lib/cli1Package manifest.
lib/crewai-core1Package manifest.
lib/crewai-tools1Package manifest.
Entry point kindCountWhere
crewai CLI commands56lib/cli/src/crewai_cli/cli.py
Release tooling console scripts5lib/devtools/pyproject.toml
Standalone scripts run as __main__7scripts/, lib/crewai-tools/src/crewai_tools/generate_tool_specs.py
Generated project templates (crew and flow)3lib/cli/src/crewai_cli/templates/
Browser event handlers in the flow visualiser18lib/crewai/src/crewai/flow/visualization/assets/interactive.js

3. Ratings

Engineering standards

Engineering standards: 7/10, Good

Weakest: duplication and dead code

Lint, format, type checks, pre-commit, CodeQL and a vulnerability scan all run in CI (.github/workflows/linter.yml, type-checker.yml, codeql.yml), against 192 groups of duplicated bodies (3,291 removable lines, mostly the CLI test suite copied between lib/cli/tests and lib/crewai/tests/cli) and 168 functions over 100 lines, the longest being setup_listeners at 904 lines in lib/crewai/src/crewai/events/event_listener.py.

Security rating

Security: 6/10, Fair

Weakest: input validation and injection

Real controls exist (an SSRF-guarding adapter in lib/crewai-tools/src/crewai_tools/security/ssrf_adapter.py, a DML gate on NL2SQL, an opt-in gate on callback deserialisation, signature-verified JWTs), but lib/crewai/src/crewai/agent/core.py:1553 unpickles trained_agents_data.pkl from the working directory on every agent task, the same directory FileWriterTool writes to by default.

Configuration and secrets rating

Configuration and secrets: 6/10, Fair

Weakest: documentation

Configuration is externalised through environment variables and CLI tokens are Fernet-encrypted in a 0700 directory (lib/crewai-core/src/crewai_core/token_manager.py), but 150 distinct variables are read with no .env.example anywhere, and lib/cli/src/crewai_cli/provider.py:162 overwrites the caller's SSL_CERT_FILE for the whole process.

Data and persistence rating

Data and persistence: 6/10, Fair

Weakest: schema evolution

Local SQLite stores use WAL and busy timeouts and checkpoints can be pruned (lib/crewai/src/crewai/state/provider/sqlite_provider.py), but every table is created with CREATE TABLE IF NOT EXISTS and nothing migrates it (lib/crewai/src/crewai/flow/persistence/sqlite.py:113), and agent memory keeps prompts and outputs in plain text with no retention limit (lib/crewai/src/crewai/memory/storage/lancedb_storage.py).

Error handling and resilience rating

Error handling and resilience: 6/10, Fair

Weakest: surfacing and swallowing

Outbound HTTP calls in the tools carry explicit timeouts and LLM clients retry twice by default (lib/crewai/src/crewai/llms/providers/openai/completion.py:385), but the evidence counts 126 empty or comment-only error handlers, a dozen in lib/cli/src/crewai_cli/crew_run_tui.py alone, so failures often vanish without a trace.

Availability and recovery rating

Availability and recovery: 5/10, Fair

Weakest: detection and alerting

As a library, most availability belongs to the host application, and flow state can be restored from checkpoints (lib/crewai/src/crewai/state/checkpoint_listener.py); but the A2A server it can expose has no health check, and the only metric in the evidence is an event sequence counter in lib/crewai/src/crewai/events/base_events.py.

Performance and scale rating

Performance and scale: 6/10, Fair

Weakest: load evidence

Async paths run end to end (kickoff_async in lib/crewai/src/crewai/flow/runtime/__init__.py) and tool results can be cached (lib/crewai/src/crewai/agents/cache/cache_handler.py), but the repository holds no benchmark, load test or stated limit, so throughput under many concurrent crews is untested.

4. Findings a quick read turned up

SeverityFindingsRatings they fall under
High1Security rating
Medium3Error handling and resilience rating, Data and persistence rating
Low8Security rating, Configuration and secrets rating, Engineering standards, Availability and recovery rating, Performance and scale rating

These counts are a floor, not a total: they are at least the findings a quick read turned up. The full report works through every rating criterion and lists each finding with its evidence.

The three that matter most

SeverityFindingRating
HighAgents unpickle a training file from the working directory on every task, a directory agent tools can write to.Security rating
MediumFailures are silently swallowed by empty or comment-only error handlers across the CLI and library.Error handling and resilience rating
MediumAgent memory holds prompts and outputs in plain text with no retention limit.Data and persistence rating

5. What the full report covers

The Full Audit and Documentation report of this repository documents:

  • 150 configuration variables, each with where it is read and what sets it
  • 8 data stores, keys and tables, with access pattern and lifetime
  • 446 external integrations, with what breaks when each is unavailable
  • 76 outbound hosts the code calls, each checked against what the code is for
  • 26 infrastructure and packaging paths probed, present or absent
  • 6 observability signals, present or absent
  • Every entry point, with where it is declared and what handles it
  • The seven ratings reasoned criterion by criterion, with the evidence for each
  • Every finding in a severity-rated risk register, with its evidence
  • A remediation roadmap: what to fix now, next and later, each with its effort
  • README corrections as a patch, and the documents linked into one set
Download