Dizhan Xue 14edd7bdb2
feat(*): run the host as an agent manager that hands work, coding included, to sub-agents (#1010)
## Summary

Raven's host acts as the user's agent manager: it delegates production
work, including small edits, to the agent whose ownership or description
fits the task -- or, on a turn bound to a Worker Table, to the worker
whose brief covers it -- keeps track of the task and its instance,
routes follow-ups back to that instance, and gathers the results for the
user. Its own tools support briefing, checking and coordination.

- Add the resident `multi-agent-coding` skill for choosing coding
agents, keeping one instance per task, passing results by file,
isolating concurrent edits in worktrees, and handing over work when an
agent exhausts its quota.
- Keep the new host role authoritative in both the DAG skill's resident
description and its full guide. A single task uses `spawn`, or a
one-node graph when only the DAG dispatch path is available.
- Rename the generic sub-agent to `Raven-General`, with `Raven` and
`raven` retained as compatibility aliases. Instance records, handle
locks, live activity, cancellation, preferences and RPC lookups use the
same canonical identity. Historical spawn and DAG readers normalize
names before liveness checks and serialization, so tasks and transcripts
join the current instance roster. Concurrent alias calls preserve both
turns instead of overwriting one conversation snapshot. The instance
registry reads former-name rows under the current name and, on its next
write, saves every loaded row under it; transcripts, direct chats and
stored graphs stay as written.
- Resolve built-in configuration aliases before merging fields:
`Raven-General` wins over `Raven`, then `raven`, regardless of file
order. Discard older rows wholesale with a warning, preventing stale
model, permission or transport settings from leaking into the selected
row.
- Resume the most recently written message state and instance log across
former names. Saved messages take precedence over an empty chat
directory; equal timestamps prefer the current name and then the most
recent former name. Existing files stay in place. Legacy spawn records,
completed DAG manifests and unfinished graphs remain readable through
the current name.
- Emit the agent roster once at the end of `spawn`'s description. The
registry places `run_subagent_dag` and then `spawn` last, including
after late MCP or session tools. The graph tool renders the roster
itself when `spawn` is unavailable. With a Worker Table bound, dispatch
guidance selects worker labels from the live enum; the roster explains
underlying capabilities. The tool descriptions and resident DAG guide
follow the same rule.
- On a turn bound to a Worker Table, such as a Persona session, `spawn`
and the DAG nodes offer worker labels alone, and a generated Persona
never includes the generic agent. The host role on such a turn routes
production work to the worker whose brief covers it, by label, and puts
work no worker covers to the user, who can add a worker or give the
conversation another Harness, instead of doing it itself or naming an
agent outside the enum. Ordinary turns render byte for byte as before. A
Harness that `load_playbook` binds mid-turn takes effect after the
identity was assembled, so that call's result carries the same routing
as this system's trusted note, after the untrusted fence. The coding
skill picks worker labels on such a turn, for its first pick and for a
quota takeover.

## Type

- [ ] Fix
- [x] Feature
- [ ] Docs
- [ ] CI / tooling
- [ ] Refactor
- [ ] Other

## Verification

Measured at this head on Linux/Python 3.12.13.

- `UV_NO_SYNC=1 uv run --no-sync pytest -q -n 12`: **33,767 passed, 178
skipped, 2 xfailed; 9 failed and 1 error.** Eight of the failures are
bound to this box and fail the same way on current main, measured the
same day: five provider tests that read the box's proxy environment, a
Node.js path case, a LibreOffice conversion, and a fixture that root can
read anyway. The ninth, with its teardown error, is the `remote_runs`
race tracked in #990; that test passed 5 of 5 runs alone, and its file
58 of 58.
- 13 new regression tests for Worker Table routing: 12 fail against the
previous head's source and all pass now; the thirteenth checks that a
Harness loaded mid-turn leaves the next turn's role plain, and passes on
both by design. The end-to-end case drives a real Persona-bound turn:
both enums offer `research-a` alone, dispatching `Raven-General` is
refused, and the role must name neither that agent nor the roster
fallback.
- Nine single-construct mutants, each applied to an isolated copy of the
tree, each fail at least one new test: the role ignoring the table, the
default path not reading the turn's table, the uncovered-work rule
removed, the generic fallback kept in the worker routing, the coding
skill reverted, and the mid-turn note never added, added on every call,
added without a dispatch path, or phrased apart from the role.
- The host identity on ordinary turns is byte-identical to the previous
head in all 16 rendered shapes: a launched sub-agent process or not,
with or without ownership lines, and every combination of the two
dispatch tools.
- Instance registry probe, run with this head and with main on one
synthetic file: reads leave the file byte-identical; one unrelated
`upsert_spawn` saves every former-name row as `Raven-General`, keeping
the more recently updated of two colliding spellings and the later one
in the file on a tie; main then finds none of those rows under `Raven`
or `raven`.
- `uv run --no-sync python scripts/coverage_gate.py --coverage-json
<coverage.json> diff --base-ref origin/main --threshold 90`, over a
coverage run of 225 test files (8,952 passed): **99.40%**, 167/168
executable changed lines covered.
- `UV_NO_SYNC=1 make lint-types lint-imports check-large-files
check-source-language COMMIT_RANGE=origin/main...HEAD`: passed; all 10
import contracts kept.
- `UV_NO_SYNC=1 make check-commits`: both commit-message validators
passed; each also refused a dangling control commit with a malformed
subject.
- `ruff check` and `ruff format --check` over the branch's Python files
passed. `node ui-tui/scripts/gen-i18n.mjs --check` passed after the
rebase onto #1015, which also changed the i18n catalogue.
- `uv run --no-sync pre-commit run --from-ref origin/main --to-ref
HEAD`: every hook passed except the whole-TUI Prettier hook, which
reports only `newInstancePicker.tsx` and `subagentsHub.tsx`; both are
byte-identical to current main.
- Live-model replay and native Windows/macOS testing were not repeated
for these fixes.

- [x] Relevant tests pass locally
- [x] Relevant lint / type checks pass locally
- [x] User-facing docs or screenshots are updated when needed

## Risk

- The host delegates work it previously performed itself, including
small scripts. Task-to-instance routing remains prompt-based; a task can
still stay with the host after it has already answered a follow-up
itself.
- In a conversation bound to a Persona or another Harness with workers,
work no worker's brief covers now waits for the user to add a worker or
change the conversation's Harness, rather than being carried out by the
host.
- Old agent spellings remain accepted. The instance registry's first
write after the upgrade, of any row for any session, saves every
former-name row in `subagent_instances.json` as `Raven-General`, and
keeps only the more recently updated row where two spellings of one
instance collide. Builds from before the rename cannot look those rows
up, so reverting the squash commit alone does not restore their instance
bindings: that needs the registry file as it stood before that first
write. Renaming rows back by hand cannot recover a colliding row the
write dropped. Transcripts, direct chats and stored graphs are not
rewritten.
- When multiple generic-agent configuration spellings coexist, only the
newest row contributes overrides. Fields supplied exclusively by an
older row are intentionally discarded.
- Moving the dispatch tools changes the tool-array cache prefix once
after upgrade.
- A plugin participant that removes `spawn` through `select_tools` can
still leave the graph tool pointing at an absent roster, because that
selection occurs after the registry's offer check. No bundled
participant does this.

- [x] Security impact considered
- [x] Backward compatibility considered
- [x] Rollback path is clear for risky changes

## Related Issues

N/A
2026-10-20 17:56:05 +08:00
..
…
…

Raven Test Suite

The test suite covers the CLI, provider routing, context assembly, memory, channels, sandbox behavior, TUI RPC contracts, and proactive engine flows.

Running Tests

Run the default suite from the repository root:

uv run pytest

For focused work, target the relevant file or marker:

uv run pytest tests/test_cli_doctor_commands.py
uv run pytest -m "not real_llm"

The suite runs on four workers by default (-n 4 in addopts), which takes it from about ten minutes to two. Two things that costs you, both undone by -n0 on the command line:

uv run pytest -n0 --pdb tests/test_cli_doctor_commands.py
  • --pdb refuses to start under the workers: --pdb is incompatible with distributing tests. There is no debugger to attach to, since the test runs in a child process.
  • Output is no longer in file order. Reading a run top to bottom, or comparing two runs line by line, wants -n0 as well.

Tests should avoid live network calls by default. When a test needs external services or a real model, guard it behind an explicit marker or environment variable so CI and local contributors get deterministic results.

Coverage Governance

Raven uses three related checks:

  • Coverage measures line and branch execution across the default Python suite. It is the broad health signal for the production raven/ package.
  • Diff coverage measures executable lines added or modified relative to the PR target branch. Comments, deleted lines, and non-executable lines are not part of its denominator. The initial threshold is 90%.
  • The ratchet compares total line and branch coverage with the audited baseline in .github/coverage-baseline.json. Neither metric may decline by more than 0.05 percentage points. This protects the established level without pretending the current project is already near 100%.

Run the same coverage command used by CI:

make coverage

PYTHON_VERSION defaults to the CI version (3.12) and is shared by all Python Make targets. It can be overridden for exploratory compatibility runs, for example make coverage PYTHON_VERSION=3.13; only the configured CI matrix version may produce an authoritative baseline.

The command prints missing lines and creates coverage.xml, coverage.json, and htmlcov/. Open htmlcov/index.html in a browser for line-by-line detail. Generated reports are ignored by git.

After running coverage, use the local gates:

make coverage-summary
make coverage-diff COVERAGE_BASE_REF=origin/main
make coverage-ratchet
make coverage-baseline-check COVERAGE_BASE_REF=origin/main

The summary prints separate total line and branch percentages plus the ten lowest-covered production files. A diff failure lists every uncovered changed line as path:line. A ratchet failure prints the current value, baseline, change in percentage points, and the files with the largest declines.

The CI unit job runs the default suite once and reuses its data for all gates. It uploads XML, JSON, and HTML reports even when tests or a gate fail. Diff coverage runs only for pull requests because a push has no PR target branch; push builds still enforce the overall ratchet.

Baseline updates

make coverage-baseline-candidate creates coverage-baseline-candidate.json from the current report. The tracked bootstrap baseline is accepted only after the introducing PR reproduces it in Python 3.12 Ubuntu CI within the ratchet tolerance. After bootstrap, the baseline must only be replaced with a candidate downloaded from a successful main CI run. Review the reference commit and totals in the candidate before replacement. Never edit the percentages by hand or lower the baseline to make a change pass. PR CI compares a proposed baseline with the target branch and rejects any decrease in either line or branch coverage.

Scope and exclusions

The default coverage suite excludes tests marked integration or e2e, plus tests/integration/, because they require real services, credentials, subprocess stacks, VMs, or network access and are not deterministic unit-CI inputs. Provider cases that require live credentials remain skipped.

All production Python under raven/ is measured except:

  • raven/__main__.py and evolver/__main__.py, which only forward their module entry points.

Type-checking-only blocks and if __name__ == "__main__" launcher blocks are excluded as non-runtime paths. No low-coverage feature module is omitted.