Report · marked 26 Aug 2026
bytedance/deer-flow
Solid foundations, held back in a few specific places.
Python · 80,878 stars · 2,419 files · branch main
76
/100
B+
Instructions
24/27Whether the repo tells an agent how to behave before it starts guessing. Losing marks on readme as an entry point.Setup
10/20Whether an agent can install the project and get it running without a human. Losing marks on deterministic install and pinned runtime version.Verification loop
21/25Whether an agent can check its own work. This is the category that most decides whether agent output is trustworthy. Losing marks on lint and format rules and static type checking.Context economy
13/20Whether the repo fits in a context window, or fights it. Losing marks on file sizes fit in context and repository weight.Navigability
10/10How quickly anything new can orient itself. Full marks.
Higher is better. Accent marks what is costing you most.
Fix these first
01
Deterministic install
Commit the lockfile your package manager generates. Without it, an agent's install and your install are different builds, and 'works on my machine' becomes unfalsifiable.
02
File sizes fit in context
A single file that fills the context window forces an agent to work from fragments, and it will confidently edit code it never saw. Splitting the worst offenders pays for itself immediately.
03
README as an entry point
Give the README an Install and a Usage heading with real commands under each. Agents pattern-match on those headings.
The full marking
Every deduction below names the file or setting it came from.
Instructions 24/27 A
Whether the repo tells an agent how to behave before it starts guessing.
✓Agent instruction file
12/12
AGENTS.md is present, which is the format the most tools read.
Also found: CLAUDE.md, Copilot instructions
✓Instruction quality
12/12
Specific enough that an agent can act on it.
14,165 characters, a workable length
names the actual commands to run
uses headings, so an agent can skim it
includes copyable code blocks
✗README as an entry point
0/3
No README. This is the first file anything reads, human or machine.
Give the README an Install and a Usage heading with real commands under each. Agents pattern-match on those headings.
Setup 10/20 C+
Whether an agent can install the project and get it running without a human.
✗Deterministic install
0/6
No lockfile. An agent installing dependencies may not get what CI got.
Commit the lockfile your package manager generates. Without it, an agent's install and your install are different builds, and 'works on my machine' becomes unfalsifiable.
✓Discoverable commands
6/6
Commands are declared where an agent will look for them.
Makefile
scripts/ directory
✓Environment config
4/4
.env.example documents what the app needs to run.
✗Pinned runtime version
0/2
No pinned runtime version.
Add a .nvmrc, .python-version, or equivalent so the agent's toolchain matches yours.
✗Reproducible environment
0/2
No container or devcontainer definition.
A Dockerfile or devcontainer removes an entire class of 'it won't install' failures.
Verification loop 21/25 A-
Whether an agent can check its own work. This is the category that most decides whether agent output is trustworthy.
✓Tests exist
8/8
797 test files against 1916 source files.
e.g. backend/tests/_agent_e2e_helpers.py, backend/tests/_replay_fixture.py, backend/tests/_router_auth_helpers.py
✓Test command is discoverable
7/7
An agent can find and run `make test`.
✓Continuous integration
4/4
14 GitHub Actions workflows define what "passing" means.
✗Lint and format rules
0/3
No linter or formatter config. Style is tribal knowledge, so agent output will drift from yours.
A formatter config is the cheapest way to stop reviewing whitespace in agent diffs. It moves style from opinion to a command.
✗Static type checking
2/3
Type checking is configured, but not in strict mode.
A type checker gives an agent an error message instead of a runtime surprise. It's the second-fastest feedback loop after the compiler.
Context economy 13/20 B
Whether the repo fits in a context window, or fights it.
✓No committed build output
6/6
No generated directories committed.
✓.gitignore hygiene
3/3
.gitignore covers 44 patterns.
✗File sizes fit in context
0/6
15 source files are over 100KB.
A single file that fills the context window forces an agent to work from fragments, and it will confidently edit code it never saw. Splitting the worst offenders pays for itself immediately.
backend/tests/test_channels.py — 402KB
backend/tests/test_client.py — 173KB
backend/tests/test_threads_router.py — 152KB
✗Repository weight
4/5
About 58.1MB checked out, which is comfortable.
Here is your AGENTS.md
Drafted from what is actually in this repository: the install
command from your lockfile, the commands you already declare, your real directory layout. Anything
marked TODO needs a person. Save it at the root as AGENTS.md.
# AGENTS.md An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours. ## Setup ``` # TODO: the install command. No lockfile was found, so this could not be inferred. ``` ## Commands ``` make help make setup make doctor make support-bundle make detect-blocking-io make config ``` ## Layout - `backend/` 1215 source files - `frontend/` 615 source files - `scripts/` 41 source files - `skills/` 23 source files - `docker/` 5 source files - `tests/` 5 source files ## Conventions - Tests live alongside the code they cover, following `backend/tests/_agent_e2e_helpers.py`. - CI defines what passing means. See `.github/workflows/backend-blocking-io-tests.yml`, and keep it green. - TODO: add the two or three conventions a newcomer always gets wrong here. ## Gotchas - `backend/tests/test_channels.py` is 402KB. It will not fit comfortably in context, so read it in parts. - Copy `.env.example` before running anything that needs configuration. - No lockfile is committed, so an install here may not match what CI produced. --- Drafted by agentreadme.com from what is in this repository. Everything marked TODO needs a human. Check it in as AGENTS.md at the root.
Open the raw markdown
or
curl -o AGENTS.md agentreadme.com/draft/bytedance/deer-flow.md
Show the mark
The badge re-checks daily, so it keeps up as the repository changes. Use mark again to force it now.
[](https://agentreadme.com/bytedance/deer-flow)Other Python repositories, marked
Think this mark is wrong?
Every deduction above names the file it came from, so this can be settled by looking. If a check missed something, that is a rule worth fixing.