Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →fedoroff-vlad/document-project
Use after the project is mapped, to write the human- and agent-facing docs: a per-module README, a root README with the module map, and an AGENTS.md spec skeleton so an agent can develop against it. Fires on: "сделай ридми по проекту", "напиши README по каждому модулю", "задокументируй проект", "нужна документация для онбординга", "write the READMEs", "document each module". Reads docs/onboarding/project-map.md; writes README.md per module + at the root, and a spec skeleton.
5 of 23 test cases passed with the skill.
Tokens, turns and time describe resource use, not a dollar price.
Lift withheld: 7 comparable cases; this report requires 8.
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
Our repo contains 45 leaf utility packages in libs/ that each contain only 10 lines of static helper code, along with 2 primary microservices. Generate the documentation hierarchy from project-map.md without creating individual README files for every minor package.
| Checked | Without skill | With skill |
|---|---|---|
| Summarizes trivial leaf packages as one-line entries within the group README rather than creating dedicated README files. | Passed | Passed |
| Creates full dedicated README files for non-trivial modules. | Missed | Passed |
The instructions your model reads when the skill loads.
The documentation phase. Input is docs/onboarding/project-map.md (from
map-project); output is a layered doc set: every module gets a README, the
root gets an overview + module map, and each service gets an AGENTS.md spec
skeleton so future work (human or agent) has intent-in-repo to build against.
The READMEs are human-facing → write them in ONBOARDING_LANG (env var,
default ru; en/other to override). The per-service AGENTS.md spec skeletons
are agent-facing → always English. In every language, keep code, commands,
identifiers, env-key names, ports and paths verbatim — translate only prose and
headings. (Full policy in the orchestrator's AGENTS.md.)
Write only what the map and the code support. A build or run step you have not
executed yet is not yet fact — tag it > ⚠️ unverified until run-guide
confirms it, then the tag comes off. Never invent an env var, a port, or a
command that is not in the repo. This keeps the docs honest across the onboarding
loop instead of accumulating plausible fiction.
A 60-module monorepo does not get 60 hand-written READMEs on the first pass — that is noise no one reads. Tier by value:
platform/,
domains/<x>/, libs/) describing that group and listing its modules.State the tiering you chose in progress.md so a resumed session continues it.
If a doc already exists (human-written or a previous pass), this is an update,
test cases passed with the skill
More skills from the same source and nearby in the registry, with their headline benchmark lift.