Port the agent-team enhancement layer onto upstream/main so the model
can spin up coordinated sub-agent teams for parallel work.
subagentModel config wiring (fix):
- Add subagent_model field to RuntimeFeatureConfig + RuntimeConfig::subagent_model()
accessor so the subagentModel setting (already validated by config_validate.rs)
is now actually read and stored.
- Agent tool's resolve_agent_model falls back to subagentModel from config
when no explicit model is passed.
Provider namespace separation:
- New 'custom-openai' provider kind with dedicated env vars
(CLAWCUSTOMOPENAI_API_KEY / CLAWCUSTOMOPENAI_BASE_URL)
- /setup wizard saves kind: 'custom-openai' for option 5
- Bare model name normalized to 'custom/' prefix to avoid proxy 404s
- Sub-agents inherit /setup-saved provider config via inject_config_as_env_fallbacks
Parallel tool execution:
- Override execute_batch to classify read-only tools as parallel-safe
and run them concurrently via std::thread::scope
- Results return in original model order
Team coordination layer:
- AgentMessage, TaskClaim, TeamStatus tools + shared mailbox directory
- Mode presets (tiny/1x ... mega/6x) + enriched TeamCreate/Agent descriptions
- Background team watcher
- /team slash command: on/off/status/toggle
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Fixes#3259: When rendering markdown in streaming mode, block-level elements
(code blocks, lists, tables, etc.) were losing trailing newlines due to
render_markdown() calling trim_end(). This caused improper spacing when chunks
were split at block boundaries.
Solution:
- Track last_output_ends_with_newline in MarkdownStreamState to detect when
previous output ended without proper spacing
- In push(): add leading newlines before block elements if previous chunk
didn't end with newlines
- In push(): restore trailing newlines after code block/table closings that
were trimmed by render_markdown()
- Add ensure_newlines() helper to prepare spacing before block starts
- Add comprehensive streaming tests that verify idempotence across all
possible chunk split positions
This ensures that streaming renders produce byte-identical output regardless
of where the input is split into chunks, which is critical for displaying
markdown as it arrives from the API while maintaining proper formatting.
Five defects found by auditing the agent stack, all of which let the harness act
confidently on a wrong picture of its own state.
1. A turn that errored discarded the record of tools that had already run. The CLI
hands run_turn a CLONE of the session; the turn appends the user message, each
assistant reply, and a tool_result per executed tool to that clone alone. The error
path dropped it while the side effects of those tools remained on disk, and the
recovery path then replayed the same user input against a session showing no sign
they had ever run — inviting a second execution of tools that are not idempotent.
The clone is now adopted before teardown, and recovery resumes the turn instead of
replaying it (new ConversationRuntime::resume_turn).
2. Every sizing guard was computed against the model NAME's context window. With a
local server behind a cloud model alias that is 200k against a much smaller
backend, which does not make the preflights inaccurate — it makes them unreachable.
CLAW_CONTEXT_WINDOW now declares the real window; model_token_limit honours it and
returns a limit for unrecognised models instead of None, max_tokens is capped to a
quarter of it, and the auto-compaction threshold derives from it (70%) rather than
defaulting to a 200k-window constant.
3. Whether to recover by discarding conversation history was decided by substring
matching on rendered error prose, including strings that ordinary transport and
decode failures also produce — so a dropped connection cost the session its
transcript. Classification now happens in the API layer, where the error still has
its type and the request is still in hand: typed overflow errors, plus ambiguous
transport failures only when the request actually filled the window. The two
copy-pasted detection blocks are gone.
4. The top-level agent loop had no iteration ceiling (usize::MAX) while spawned
subagents were bounded at 32. It now gets the same bound, overridable with
CLAW_MAX_TURN_ITERATIONS.
5. Auto-continue nudges were injected as USER turns, so harness-authored text was
persisted, re-sent on every later request, and folded into compaction summaries as
something the user had said. They are recorded as system turns now; the wire
payload is unchanged.
Also replaces the repository's own self-host section, which described a translator
and a serving stack no longer in the request path. That file is folded into the
system prompt verbatim, so it was briefing the model with an architecture that does
not exist.
(cherry picked from commit c8834cb4d1a3372eb70241429f641162a420297e)
The suite asserted against whatever the developer's box happened to be
configured as. Two independent leaks, both surfaced by a real `~/.claw`:
1. `parse_args` resolves defaults — notably `permissions.defaultMode` —
through the user-scope config, which `default_config_home()` reads from
`CLAW_CONFIG_HOME` falling back to `$HOME/.claw`. Nothing overrode it, so
a `defaultMode` of `dontAsk` on the host turned 12 assertions from
`WorkspaceWrite` into `DangerFullAccess`. `env_lock()` now returns an
`EnvGuard` that points `CLAW_CONFIG_HOME` at a fresh empty directory and
restores the previous value on drop, so "no user config" is the state
under test. All 42 call sites already bind the guard, so this fixes them
at once; tests that want a populated config home can still set the var
themselves afterwards. `removed_login_and_logout_subcommands_error_helpfully`
asserted a Default-sourced permission mode without taking the guard at
all, and now does.
2. `direct_resume_safe_slash_commands_route_to_local_json_actions_831`
hardcoded `/sandbox` => "warn", which is only true where namespace
isolation is unsupported and just the filesystem sandbox is active
(#731). On a host with working user namespaces the correct answer is
"ok". That entry now accepts either; the test is about #831 routing —
local JSON action instead of `interactive_only` — not about the kernel
under the runner.
Verified with the real ~/.claw/settings.json in place (defaultMode=dontAsk):
cargo test --workspace 0 failures, clippy --workspace --all-targets
-D warnings 0 findings, scripts/fmt.sh clean.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q9McEEH32ZtyM62GSXVSyJ
(cherry picked from commit 67aa4530c22f6fdf95a29d6a46992af137134c67)
`cargo clippy --workspace --all-targets -- -D warnings` is this repo's own
documented verification command, and it could not run to completion. Earlier
compile failures were masking a further 48 lints; with those cleared the
workspace now passes with zero findings.
Real fixes, not suppressions:
- status_json_value took 8 positional parameters, four of them a tail of
`Option<&_>` that most callers passed as None — trivially transposable.
Grouped into `StatusJsonExtras` with a Default impl; all 7 call sites now
name only the fields they actually set.
- extract_help_metadata's 8-tuple return is now a documented `HelpMetadata`
alias.
- the auto-compact loop indexed preserve_schedule by its loop counter;
it now iterates the array directly.
- a sandbox status chain had two identical arms; merged the condition.
- a filter_map whose every arm returned Some is now a map.
- 33 unused stdout/stderr bindings in the output-format contract tests are
prefixed with `_`, and 9 `map_or(false, ..)` are now `is_some_and(..)`.
NOTE: several `--bin claw` tests (parse_args defaults) read the *real*
`~/.claw/settings.json` rather than an isolated config home, so they fail on
any machine whose user config sets `permissions.defaultMode`. With that file
moved aside the binary's suite is 207/207. That isolation gap is pre-existing
and left for a separate change.
Verified: clippy workspace/all-targets/-D warnings reports 0; cargo test
--workspace passes with the user config neutralised; scripts/fmt.sh clean.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q9McEEH32ZtyM62GSXVSyJ
(cherry picked from commit 03c8a8fd7ba6e405a8217d00d3766c1c6367e527)
`ApiError` is the `Err` half of nearly every `Result` in the api crate, and
its `Api` variant carried ~139 bytes inline (8 fields, four of them
`Option<String>`). That tripped clippy::result_large_err on 35 signatures
and, with -D warnings, made the whole crate unbuildable under the repo's
own documented verification command.
Move the payload into `ApiErrorDetails` behind a `Box`:
- `ApiError::Api(Box<ApiErrorDetails>)`, with `ApiError::api(..)` to
construct and `api_details()` to borrow
- `ApiErrorDetails` re-exported from the crate root
- adding a field here can no longer re-inflate every signature
This also simplifies `enrich_bearer_auth_error`, which previously
destructured and rebuilt all eight fields at five separate return points
just to append a hint; it now mutates the boxed payload in place.
Also clears the remaining pre-existing lints that blocked -D warnings:
unused test bindings, a redundant `use serde_json`, and an explicit
`.into_iter()` in claw-rag-service.
Verified: cargo test --workspace passes, 0 failures.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q9McEEH32ZtyM62GSXVSyJ
(cherry picked from commit e9210bd75eb76e5d88b2ae797f537102bcabd41d)
refactor(cli): let the emitter report the output tail it leaves behind
Two places believed they owned terminal output: consume_stream streams
assistant deltas live while emit_output = true, and run_turn decided
whether to emit a separating newline before spinner.finish by guessing
from final_assistant_text(&summary).
That guess agrees with reality only because every *other* writer happens to
terminate its output with a newline — tool results go through
stream_markdown (which appends one), thinking summaries are written with a
trailing newline, and tool-call headers use writeln! — so the assistant
text path is the only one that can leave a partial line. That is a property
of the current renderers rather than of the streaming protocol, and it is
not enforced anywhere.
TerminalTail is state the emitter records into as it writes: both live
writers (AnthropicRuntimeClient::consume_stream and the tool-result blocks
in CliToolExecutor::execute) now push bytes through a TailTrackingWriter
that reports the last byte written, and the caller that ends a turn asks
where the emitter stopped instead of inferring it.
The heuristic also left the failure arm unguarded. A turn that streamed text
and then errored called spinner.fail on a partial line, which the line
clear then erased; that arm now carries the same guard.
Also gate the animated spinner.tick frame on is_terminal(). That frame is
pure cursor control (save, rewrite line, restore) and emits no newline, so
down a pipe it leaves an unterminated banner that the first streamed delta
then continues on the same physical line.
—
*[repo owner's gaebal-gajae (clawdbot) 🦞]*
Route groq-prefixed models through the OpenAI-compatible client,
apply Groq token-limit metadata, and skip local tool injection for
compound models.
Add regression coverage for routing, token limits, and compound
local-tool behavior.
Co-authored-by: Oz <oz-agent@warp.dev>
feat: add persistent auto-memory system
Implement a file-based persistent memory system that allows the AI to retain knowledge about users, feedback, projects, and references across sessions.
## Implementation Details
- MemoryStore data layer with FNV-1a workspace fingerprinting
- MemoryRead/MemoryWrite tools with security hardening (path traversal, symlink, absolute path, and size limit protections)
- System prompt injection of MEMORY.md index with prompt injection mitigation (fenced code block + trust-lowering notice)
- autoMemoryEnabled config option (defaults to true)
- Doctor health check for auto-memory state
- /memory command enhancement showing persistent memory info
## Verification
✓ 1023 lib tests pass
✓ Output format contract tests pass
✓ Release build succeeds
✓ Security hardening verified
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* fix(runtime): prompt permission mode never invokes the prompter
PermissionMode derives Ord from declaration order, placing Prompt (and Allow)
above the real privilege ladder (ReadOnly < WorkspaceWrite < DangerFullAccess).
In PermissionPolicy::authorize_with_context the ladder check
'current_mode >= required_mode' therefore evaluates true for every tool when the
active mode is Prompt (3 >= {0,1,2}), returning Allow and making the subsequent
'if current_mode == PermissionMode::Prompt' branch that routes to prompt_or_deny
unreachable.
Effect: a session configured with the 'prompt' permission mode silently
auto-allows every tool and never calls the prompter (reached via
ConversationRuntime::run_turn with an interactive prompter), defeating the mode's
entire purpose. The sibling PermissionEnforcer already special-cases Prompt
before its own 'active_mode >= required_mode' check, confirming Prompt is not
meant to be part of the ordered ladder.
Fix: exclude Prompt from the ordered-ladder comparison so Prompt mode falls
through to the interactive prompt path. Allow remains handled by its explicit
equality check. The same comparison appears twice in authorize_with_context —
once in the shared tail and once inside the 'Some(PermissionOverride::Allow)'
arm, which is reached when a hook emits permissionDecision "allow" — so both
copies carry the guard; guarding only the tail still left a hook able to
auto-allow every tool in Prompt mode without ever consulting the prompter.
Adds regression tests covering Prompt mode with a prompter, without a prompter,
and with a hook allow override.
* refactor(runtime): make the permission privilege ladder an explicit rank
`PermissionMode` derived `PartialOrd`/`Ord`, so the ladder deciding whether a
tool runs without asking was declaration order. A variant inserted in the middle
of the enum re-ranked it silently, and the difference between "this tool may
run" and "this tool may run after a prompt" was a property of where an `enum`
line happened to sit. That is how `Prompt` came to outrank `DangerFullAccess`,
satisfy `current_mode >= required_mode` for every tool, and skip the prompter.
Replace the derive with `privilege_rank`, an exhaustive `match` naming every
variant, so inserting one is a build failure rather than a silent re-ranking.
`Ord` is deliberately not derived, so a stray `>=` on two modes no longer
compiles; the two comparisons outside this module now say which rank they read
(`PermissionEnforcer::check_with_required_mode` and the `claw doctor` permission
check). Both keep their previous answers, since `privilege_rank` reproduces the
old declaration order exactly, including `Prompt` and `Allow` above the tiers.
`Prompt` and `Allow` are modes of operation, not privilege tiers, so they are
never ranked *against* a tool requirement. The new `covers` predicate states
that by name rather than by number: `Allow` always covers, `Prompt` never does
(it asks), and only the three tiers compare ranks.
That also collapses the duplicated ladder in `authorize_with_context`. The
`Some(PermissionOverride::Allow)` arm carried a second, identical copy of the
ask-rule check and the ladder; it now falls through to the single copy below,
so a hook `allow` is an override rather than a bypass and the two copies can no
longer drift apart.
Behaviour is unchanged for every (mode, requirement, rule, override)
combination. The tests pin the `Prompt` routes — plain and hook-`allow` — across
all three tiers, plus the rank ordering itself.
Root knowledge base plus complexity-scored subdirectory files for the
rust/ workspace, its five highest-mass crates (runtime, rusty-claude-cli,
api, tools, commands, plugins), and the src/ Python porting workspace.
Generated via init-deep: 13 parallel explore agents, LSP/ast-grep code
map, centrality-scored placement. Snapshot in .omo/init-deep.json (local).
- OLLAMA_HOST takes priority over OPENAI_BASE_URL for local Ollama instances
- No API key required; placeholder token used for Authorization header
- Model names like 'qwen3:8b' bypass strict provider/model syntax validation
- detect_provider_kind() checks OLLAMA_HOST first in routing cascade
- ProviderClient dispatch uses from_ollama_env() when OLLAMA_HOST is set
- Updated USAGE.md and docs with OLLAMA_HOST as preferred env var
- Added OLLAMA_CONFIG constant and from_ollama_env() to openai_compat
- Added test_ollama_host_bypasses_model_validation unit test
- Supersedes PR #3213 (which had a duplicate if-let bug in mod.rs)
The setup wizard was merged in PR #3017 but was orphaned -- it was not
declared as a module in main.rs, making it unreachable. Additionally,
the setup_wizard.rs imports RuntimeProviderConfig which did not exist
on upstream/main. This commit makes the wizard accessible and adds the
necessary RuntimeProviderConfig type.
Changes:
- Add RuntimeProviderConfig struct to runtime/src/config.rs with
kind(), api_key(), base_url(), model() accessors.
- Add parse_optional_provider_config() to parse the provider object
from merged settings JSON.
- Add provider() method to RuntimeConfig and RuntimeFeatureConfig.
- Export RuntimeProviderConfig, save_user_provider_settings,
clear_user_provider_settings, and default_config_home from runtime
crate public API (runtime/src/lib.rs).
- Add mod setup_wizard to rusty-claude-cli/src/main.rs.
- Add claw setup CLI subcommand.
- Add /setup slash command.
- Add Setup variant to SlashCommand enum.
- Add Setup to LocalHelpTopic enum.
- Add setup to diagnostic subcommand matching.
- Add subagentModel to TOP_LEVEL_FIELDS in config_validate.rs.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
SessionLifecycleSummary now collects all matching tmux panes into an
all_panes field and includes them in the JSON output. Previously the
status command returned on the first non-idle pane, losing all other
active panes in the same workspace/session.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
/resume mode /clear now preserves the original session_id instead of
generating a new one. This prevents the filename/meta-header divergence
where /session list reported an id that --resume couldn't find.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
/session switch and /session fork now return structured JSON with
kind:error, error_kind:unsupported_resumed_command, and actionable
hint instead of a raw error string that resume callers couldn't parse.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
Add friendly_cwd() helper that strips /private prefix on macOS so
status, doctor, and diff JSON output matches user-visible invocation
cwd instead of the canonicalized /private/tmp path.
Applied to status_context() and render_doctor_report().
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
/model was in the unsupported resume group but is a read-only command
that can safely return model configuration. Now returns default_model,
configured_model, resolved_model, and requested_model in JSON mode.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
--compact was silently ignored when used with non-prompt commands like
claw --compact status or claw --compact config. Now returns a typed
error with guidance on proper usage.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
/tasks was marked resume_supported but rejected in resume mode.
Now returns a structured JSON response with kind:'tasks' and note
that background tasks are only available in the interactive REPL.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
- /config help now returns available_sections array and loaded_keys
count instead of treating 'help' as an unsupported section
- Updated test to exclude 'help' from unsupported sections test
- Added new test config_help_returns_structured_section_list_344
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
/config help now returns a list of available config sections in both
text and JSON mode instead of treating 'help' as an unsupported section.
Text mode shows available sections with descriptions.
JSON mode returns available_sections array with loaded_keys count.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
Top-level 'claw --output-format json help' now includes a 'commands'
array with name, summary, and resume_supported for each registered
slash command, plus 'total_commands' count.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
Add GitOperation enum to detect mid-operation git states from the
branch header in git status --short --branch output.
- Rebase: 'rebasing ...' in branch header
- Merge: '[merge-in-progress]' tag
- Cherry-pick: 'cherry-pick-in-progress' tag
- Bisect: 'bisect-in-progress' tag
Operation state appears in:
- status text: 'rebase-in-progress, dirty · 3 files · ...'
- status JSON: 'git_operation' field (null when no operation)
- git_state headline includes operation prefix
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
--base-commit now rejects non-hex strings and strings outside 7-64
character range. Matches the pattern used by --reasoning-effort.
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
The --cwd validation added in #99 rejects non-existent paths, but the
test used /tmp/project which doesn't exist on CI Linux runners. Changed
to /tmp which exists everywhere.
Also marks ROADMAP #123 DONE (--allowedTools normalization verified).
Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
TheArchitectit
Bellman
clawdbot
DaBear987
reacechalmers
黄云龙
AbdullahFageehandOz
Ming LiuandClaude Opus 4.6
gaebal-gajae
YeonGyu-Kim
nankingjing
Aneesh
User
bellman
Your NameandClaude Opus 4.8
