Commit Graph
14 Commits
Author SHA1 Message Date
黄云龙 86428abce6 refactor(runtime): make the permission privilege ladder an explicit rank (#3303)
* 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.
2026-10-13 07:04:44 +09:00
bellman aca6584fd5 fix: normalize permission rule tool names to lowercase (#94)
PermissionRule::parse now normalizes tool_name to lowercase, matching
the runtime convention. Previously "Bash(rm:*)" would never match
because the runtime tool name is lowercase "bash". Same fix applied
to denied_tools list.

Generated with https://github.com/Yeachan-Heo/gajae-code
Co-authored-by: Gajae Code <dev@gajae-code.com>
2026-06-05 09:53:26 +09:00
YeonGyu-Kim 495e7a015c fix: remove stale retry_after field, Team variant, config_load_error_kind, denied_tools initializer errors
- Remove retry_after: None from ApiError::Api structs in openai_compat.rs (field was removed)
- Remove SlashCommand::Team parse arm (variant was removed from enum)
- Add config_load_error_kind: None to doctor path StatusContext initializer
- Add Thinking arm to all ContentBlock match blocks in trident.rs
- Remove cargo fmt drift across commands, config, compact, tools, trident
2026-05-25 12:01:09 +09:00
bellman 7214573f35 Keep approval token contracts in their own runtime module
Constraint: G004 task 3 now owns approval-token contracts through rust/crates/runtime/src/approval_tokens.rs, while auto-integration left a duplicate unused copy in permissions.rs.\nRejected: suppressing dead-code warnings | the duplicate implementation was obsolete after the dedicated module landed.\nConfidence: high\nScope-risk: narrow\nDirective: Keep permission-mode authorization in permissions.rs and approval-token policy handoff in approval_tokens.rs.\nTested: cargo fmt --manifest-path rust/Cargo.toml --all -- --check; cargo check --manifest-path rust/Cargo.toml -p runtime; cargo test --manifest-path rust/Cargo.toml -p runtime approval_token -- --nocapture; cargo test --manifest-path rust/Cargo.toml -p runtime --test g004_conformance -- --nocapture\nNot-tested: full workspace test suite; G004 tasks 2/4/5 remain non-terminal.\n\nCo-authored-by: OmX <omx@oh-my-codex.dev>
2026-05-14 18:11:20 +09:00
bellman 5cebdd999d omx(team): auto-checkpoint worker-2 [3] 2026-05-14 18:07:05 +09:00
bellman e34209ff7f omx(team): auto-checkpoint worker-2 [3] 2026-05-14 18:07:00 +09:00
bellman ff37d395bb Stabilize G004 contract integration after worker merges
Constraint: G004 worker integrations introduced unparseable approval-token tests and a conformance path bug that blocked leader verification.\nRejected: waiting for another auto-integration cycle | local leader verification had exact parse and fixture failures to repair safely.\nConfidence: high\nScope-risk: moderate\nDirective: Keep approval-token regression tests in cfg(test) modules or integration tests, never inside type definitions.\nTested: cargo fmt --manifest-path rust/Cargo.toml --all -- --check; cargo check --manifest-path rust/Cargo.toml -p runtime; cargo test --manifest-path rust/Cargo.toml -p runtime approval_token -- --nocapture; cargo test --manifest-path rust/Cargo.toml -p runtime --test g004_conformance -- --nocapture; python3 .github/scripts/check_doc_source_of_truth.py\nNot-tested: full workspace test suite; remaining G004 tasks 1-5 still non-terminal.\n\nCo-authored-by: OmX <omx@oh-my-codex.dev>
2026-05-14 18:06:14 +09:00
bellman f4e08d0ecf omx(team): auto-checkpoint worker-2 [3] 2026-05-14 17:58:46 +09:00
bellman 16d6525de4 omx(team): auto-checkpoint worker-2 [3] 2026-05-14 17:57:59 +09:00
Yeachan-Heo 22ad54c08e docs: describe the runtime public API surface
This adds crate-level and type-level Rustdoc to the runtime crate's core exported types so downstream crates and contributors can understand the session, prompt, permission, OAuth, usage, and tool I/O primitives without spelunking every implementation file.

Constraint: The docs pass needed to stay focused on public runtime types without changing behavior
Rejected: Add blanket docs to every public item in one sweep | larger churn than needed for a targeted docs pass
Confidence: high
Scope-risk: narrow
Reversibility: clean
Directive: When exporting new runtime primitives from lib.rs, add a short Rustdoc summary in the defining module at the same time
Tested: cargo build --workspace; cargo test --workspace
Not-tested: rustdoc HTML rendering beyond  doc-test coverage
2026-04-04 15:23:29 +00:00
Yeachan-Heo eb89fc95e7 wip: hook-pipeline progress 2026-04-01 04:30:25 +00:00
Yeachan-Heo 387a8bb13f feat: git integration, sandbox isolation, init command (merged from rcc branches) 2026-04-01 01:23:47 +00:00
Yeachan-Heo e2f061fd08 Enforce tool permissions before execution
The Rust CLI/runtime now models permissions as ordered access levels, derives tool requirements from the shared tool specs, and prompts REPL users before one-off danger-full-access escalations from workspace-write sessions. This also wires explicit --permission-mode parsing and makes /permissions operate on the live session state instead of an implicit env-derived default.

Constraint: Must preserve the existing three user-facing modes read-only, workspace-write, and danger-full-access

Constraint: Must avoid new dependencies and keep enforcement inside the existing runtime/tool plumbing

Rejected: Keep the old Allow/Deny/Prompt policy model | could not represent ordered tool requirements across the CLI surface

Rejected: Continue sourcing live session mode solely from RUSTY_CLAUDE_PERMISSION_MODE | /permissions would not reliably reflect the current session state

Confidence: high

Scope-risk: moderate

Reversibility: clean

Directive: Add required_permission entries for new tools before exposing them to the runtime

Tested: cargo fmt; cargo clippy --workspace --all-targets -- -D warnings; cargo test -q

Not-tested: Manual interactive REPL approval flow in a live Anthropic session
2026-04-01 00:06:15 +00:00
Yeachan-Heo 44e4758078 feat: Rust port of Claude Code CLI
Crates:
- api: Anthropic Messages API client with SSE streaming
- tools: Claude-compatible tool implementations (Bash, Read, Write, Edit, Glob, Grep + extended suite)
- runtime: conversation loop, session persistence, permissions, system prompt builder
- rusty-claude-cli: terminal UI with markdown rendering, syntax highlighting, spinners
- commands: subcommand definitions
- compat-harness: upstream TS parity verification

All crates pass cargo fmt/clippy/test.
2026-03-31 17:43:09 +00:00