Commit graph

43 commits

Author SHA1 Message Date
Chummy
50fd5b81e1 fix(test): stabilize cron output capture and clippy cleanups 2026-02-18 20:29:26 +08:00
Chummy
483acccdb7 feat(memory): add configurable postgres storage backend 2026-02-18 20:29:26 +08:00
Chummy
d42cb1e906 fix(auth): rebase PR #200 onto main and restore auth CLI flow 2026-02-18 12:57:44 +08:00
Codex
96109d46d1 Fix pending OAuth verifier storage and account id fallback 2026-02-18 12:57:44 +08:00
Codex
007368d586 feat(auth): add subscription auth profiles and codex/claude flows 2026-02-18 12:57:44 +08:00
Chummy
e85418eda4 chore(ci): align formatting and clippy output for gates 2026-02-18 00:50:51 +08:00
Chummy
ce23cbaeea fix(cli): harden providers listing and keep provider map aligned 2026-02-18 00:50:51 +08:00
reidliu41
feaa4aba60 feat(cli): add zeroclaw providers command to list supported providers
- Add `zeroclaw providers` CLI command that lists all 28 supported AI providers
- Each entry shows: config ID, display name, local/cloud tag, active marker, and aliases
- Also shows `custom:<URL>` and `anthropic-custom:<URL>` escape hatches at the bottom

Previously users had no way to discover available providers without reading source code. The
unknown-provider error message suggests `run zeroclaw onboard --interactive` but doesn't list
options. This command gives immediate visibility.
2026-02-18 00:50:51 +08:00
stawky
ab561baa97 feat(approval): interactive approval workflow for supervised mode (#215)
- Add auto_approve / always_ask fields to AutonomyConfig
- New src/approval/ module: ApprovalManager with session-scoped allowlist,
  ApprovalRequest/Response types, audit logging, CLI interactive prompt
- Insert approval hook in agent_turn before tool execution
- Non-CLI channels auto-approve; CLI shows Y/N/A prompt
- Skip approval for read-only tools (file_read, memory_recall) by default
- 15 unit tests covering all approval logic
2026-02-17 23:06:12 +08:00
leon
fa94117269 feat(telegram): add operator bind command for unauthorized users 2026-02-17 22:48:40 +08:00
Alex Gorevski
529a3d0242
fix(cli): respect config gateway.port and gateway.host for Gateway/Daemon commands (#456)
The CLI --port and --host args had hardcoded defaults (8080, 127.0.0.1)
that always overrode the user's config.toml [gateway] settings (port=3000,
host=127.0.0.1). Changed both args to Option types and fall back to
config.gateway.port / config.gateway.host when not explicitly provided.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-02-17 08:10:32 -05:00
Chummy
42fa802bad fix(ollama): sanitize provider payload logging 2026-02-17 18:48:45 +08:00
Kieran
b828873426 feat: accept RUST_LOG env filter 2026-02-17 18:48:45 +08:00
mai1015
fb2d1cea0b Implement cron job management tools and types
- Added `JobType`, `SessionTarget`, `Schedule`, `DeliveryConfig`, `CronJob`, `CronRun`, and `CronJobPatch` types in `src/cron/types.rs` for cron job configuration and management.
- Introduced `CronAddTool`, `CronListTool`, `CronRemoveTool`, `CronRunTool`, `CronRunsTool`, and `CronUpdateTool` in `src/tools` for adding, listing, removing, running, and updating cron jobs.
- Updated the `run` function in `src/daemon/mod.rs` to conditionally start the scheduler based on the cron configuration.
- Modified command-line argument parsing in `src/lib.rs` and `src/main.rs` to support new cron job commands.
- Enhanced the onboarding wizard in `src/onboard/wizard.rs` to include cron configuration.
- Added tests for cron job tools to ensure functionality and error handling.
2026-02-17 17:06:28 +08:00
Chummy
e197cc5b04 fix(onboard,anthropic): stabilize oauth setup-token flow and model defaults
- fix onboard command ownership handling before spawn_blocking

- restore memory helper imports in wizard to resolve build regression

- centralize Anthropic OAuth beta header in apply_auth for all request paths

- correct OpenRouter Anthropic Sonnet 4.5 model ID format

- add regression tests for auth headers and curated model IDs
2026-02-17 16:15:38 +08:00
Pedro
045ead628a feat(onboard): add Anthropic OAuth setup-token support and update models
Enable Pro/Max subscription users to authenticate via OAuth setup-tokens
(sk-ant-oat01-*) by sending the required anthropic-beta: oauth-2025-04-20
header alongside Bearer auth. Update curated model list to latest
(Opus 4.6, Sonnet 4.5, Haiku 4.5) and fix Tokio runtime panic in
onboard wizard by wrapping blocking calls in spawn_blocking.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-17 16:15:38 +08:00
Argenis
b2facc7526
fix(cli): respect config default_temperature
Fixes #452 - CLI now respects config.default_temperature

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-16 20:08:00 -05:00
ehu shubham shaw
de3ec87d16
Ehu shubham shaw contribution --> Hardware support (#306)
* feat: add ZeroClaw firmware for ESP32 and Nucleo

* Introduced new firmware for ZeroClaw on ESP32 and Nucleo-F401RE, enabling JSON-over-serial communication for GPIO control.
* Added `zeroclaw-esp32` with support for commands like `gpio_read` and `gpio_write`, along with capabilities reporting.
* Implemented `zeroclaw-nucleo` firmware with similar functionality for STM32, ensuring compatibility with existing ZeroClaw protocols.
* Updated `.gitignore` to include new firmware targets and added necessary dependencies in `Cargo.toml` for both platforms.
* Created README files for both firmware projects detailing setup, build, and usage instructions.

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

* feat: enhance hardware peripheral support and documentation

- Added `Peripheral` trait implementation in `src/peripherals/` to manage hardware boards (STM32, RPi GPIO).
- Updated `AGENTS.md` to include new extension points for peripherals and their configuration.
- Introduced comprehensive documentation for adding boards and tools, including a quick start guide and supported boards.
- Enhanced `Cargo.toml` to include optional dependencies for PDF extraction and peripheral support.
- Created new datasheets for Arduino Uno, ESP32, and Nucleo-F401RE, detailing pin aliases and GPIO usage.
- Implemented new tools for hardware memory reading and board information retrieval in the agent loop.

This update significantly improves the integration and usability of hardware peripherals within the ZeroClaw framework.

* feat: add ZeroClaw firmware for ESP32 and Nucleo

* Introduced new firmware for ZeroClaw on ESP32 and Nucleo-F401RE, enabling JSON-over-serial communication for GPIO control.
* Added `zeroclaw-esp32` with support for commands like `gpio_read` and `gpio_write`, along with capabilities reporting.
* Implemented `zeroclaw-nucleo` firmware with similar functionality for STM32, ensuring compatibility with existing ZeroClaw protocols.
* Updated `.gitignore` to include new firmware targets and added necessary dependencies in `Cargo.toml` for both platforms.
* Created README files for both firmware projects detailing setup, build, and usage instructions.

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

* feat: enhance hardware peripheral support and documentation

- Added `Peripheral` trait implementation in `src/peripherals/` to manage hardware boards (STM32, RPi GPIO).
- Updated `AGENTS.md` to include new extension points for peripherals and their configuration.
- Introduced comprehensive documentation for adding boards and tools, including a quick start guide and supported boards.
- Enhanced `Cargo.toml` to include optional dependencies for PDF extraction and peripheral support.
- Created new datasheets for Arduino Uno, ESP32, and Nucleo-F401RE, detailing pin aliases and GPIO usage.
- Implemented new tools for hardware memory reading and board information retrieval in the agent loop.

This update significantly improves the integration and usability of hardware peripherals within the ZeroClaw framework.

* feat: Introduce hardware auto-discovery and expanded configuration options for agents, hardware, and security.

* chore: update dependencies and improve probe-rs integration

- Updated `Cargo.lock` to remove specific version constraints for several dependencies, including `zerocopy`, `syn`, and `strsim`, allowing for more flexibility in version resolution.
- Upgraded `bincode` and `bitfield` to their latest versions, enhancing serialization and memory management capabilities.
- Updated `Cargo.toml` to reflect the new version of `probe-rs` from `0.24` to `0.30`, improving hardware probing functionality.
- Refactored code in `src/hardware` and `src/tools` to utilize the new `SessionConfig` for session management in `probe-rs`, ensuring better compatibility and performance.
- Cleaned up documentation in `docs/datasheets/nucleo-f401re.md` by removing unnecessary lines.

* fix: apply cargo fmt

* docs: add hardware architecture diagram.

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-16 11:40:10 -05:00
Chummy
53844f7207
feat(memory): lucid memory integration with optional backends (#285) 2026-02-17 00:31:50 +08:00
fettpl
60e72a6ed5 fix(main): remove duplicate ModelCommands enum definition
A duplicate ModelCommands enum was introduced in a recent merge,
causing E0119/E0428 compile errors on CI (Rust 1.92).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-16 17:00:10 +01:00
Chummy
a403b5f5b1
feat(onboard): add provider model refresh command with TTL cache (#323) 2026-02-16 23:40:33 +08:00
Chummy
80da3e64e9
feat: unify scheduled tasks from #337 and #338 with security-first integration
Unifies scheduled task capabilities and consolidates overlapping implementations from #337 and #338 into a single security-first integration path.\n\nCo-authored-by: Edvard <ecschoye@stud.ntnu.no>\nCo-authored-by: stawky <stakeswky@gmail.com>
2026-02-16 23:38:29 +08:00
Chummy
c842ece12c
feat(onboard): refresh model discovery and canonicalize provider aliases (#341)
* feat(onboard): add model refresh command with ttl cache

* fix(onboard): refresh curated models and canonicalize provider aliases

* fix(channels): align agent_turn call signature

* fix(channels): call run_tool_call_loop for stable channel runtime
2026-02-16 22:32:30 +08:00
mai1015
50f508766f
feat: add verbose logging and complete observability (#251) 2026-02-16 05:59:07 -05:00
Argenis
0383a82a6f
feat(security): Add Phase 1 security features
* test: add comprehensive recovery tests for agent loop

Add recovery test coverage for all edge cases and failure scenarios
in the agentic loop, addressing the missing test coverage for
recovery use cases.

Tool Call Parsing Edge Cases:
- Empty tool_result tags
- Empty tool_calls arrays
- Whitespace-only tool names
- Empty string arguments

History Management:
- Trimming without system prompt
- Role ordering consistency after trim
- Only system prompt edge case

Arguments Parsing:
- Invalid JSON string fallback
- None arguments handling
- Null value handling

JSON Extraction:
- Empty input handling
- Whitespace only input
- Multiple JSON objects
- JSON arrays

Tool Call Value Parsing:
- Missing name field
- Non-OpenAI format
- Empty tool_calls array
- Missing tool_calls field fallback
- Top-level array format

Constants Validation:
- MAX_TOOL_ITERATIONS bounds (prevent runaway loops)
- MAX_HISTORY_MESSAGES bounds (prevent memory bloat)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat(security): Add Phase 1 security features - sandboxing, resource limits, audit logging

Phase 1 security enhancements with zero impact on the quick setup wizard:
-  Pluggable sandbox trait system (traits.rs)
-  Landlock sandbox support (Linux kernel 5.13+)
-  Firejail sandbox support (Linux user-space)
-  Bubblewrap sandbox support (Linux/macOS user namespaces)
-  Docker sandbox support (container isolation)
-  No-op fallback (application-layer security only)
-  Auto-detection logic (detect.rs)
-  Audit logging with HMAC signing support (audit.rs)
-  SecurityConfig schema (SandboxConfig, ResourceLimitsConfig, AuditConfig)
-  Feature-gated implementation (sandbox-landlock, sandbox-bubblewrap)
-  1,265 tests passing

Key design principles:
- Silent auto-detection: no new prompts in wizard
- Graceful degradation: works on all platforms
- Feature flags: zero overhead when disabled
- Pluggable architecture: swap sandbox backends via config
- Backward compatible: existing configs work unchanged

Config usage:
```toml
[security.sandbox]
enabled = false  # Explicitly disable
backend = "auto"  # auto, landlock, firejail, bubblewrap, docker, none

[security.resources]
max_memory_mb = 512
max_cpu_time_seconds = 60

[security.audit]
enabled = true
log_path = "audit.log"
sign_events = false
```

Security documentation:
- docs/sandboxing.md: Sandbox implementation strategies
- docs/resource-limits.md: Resource limit approaches
- docs/audit-logging.md: Audit logging specification
- docs/security-roadmap.md: 3-phase implementation plan
- docs/frictionless-security.md: Zero-impact wizard design
- docs/agnostic-security.md: Platform/hardware agnostic approach

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-16 04:14:16 -05:00
Vernon Stinebaker
d5e8fc1652
fix(config): apply env overrides at runtime and fix Docker compose defaults (#279)
- Call apply_env_overrides() after Config::load_or_init() in main.rs so
  environment variables (API_KEY, PROVIDER, ZEROCLAW_GATEWAY_PORT, etc.)
  are actually applied at runtime, not just in tests
- Add ZEROCLAW_ALLOW_PUBLIC_BIND env var support for gateway bind policy
- Fix docker-compose.yml: correct volume path (/zeroclaw-data not /data),
  add ZEROCLAW_ALLOW_PUBLIC_BIND=true for container networking, make host
  port configurable via HOST_PORT env var
- Add docker-compose.override.yml to .gitignore for local dev overrides
2026-02-16 02:12:49 -05:00
Chummy
e04e7191ac
fix(agent): robust tool-call parsing for noisy model outputs
Improve tool-call parsing to handle noisy local-model outputs (markdown fenced JSON, conversational wrappers, and raw JSON tool objects) and add regression coverage for these cases.

Also sync rustfmt-required formatting and align crate-level clippy allow-list with Rust 1.92 CI pedantic checks so required lint gates pass consistently.

Co-authored-by: chumyin <chumyin@users.noreply.github.com>
Co-authored-by: argenis de la rosa <theonlyhennygod@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-15 23:21:26 -05:00
Argenis
dca95cac7a
fix: add channel message timeouts, Telegram fallback, and fix identity/observer tests
Closes #184
2026-02-15 12:31:40 -05:00
Argenis
128b30cdf1
fix: install default Rustls crypto provider to prevent TLS initialization error
Install ring-based crypto provider at startup to fix Rustls TLS initialization error
2026-02-15 11:10:28 -05:00
Argenis
35b63d6b12
feat: SkillForge — automated skill discovery, evaluation & integration engine (#144)
* feat: add SkillForge — automated skill discovery, evaluation, and integration engine

SkillForge adds a 3-stage pipeline for autonomous skill management:

- Scout: discovers candidate skills from GitHub (extensible to ClawHub, HuggingFace)
- Evaluate: scores candidates on compatibility, quality, and security (weighted 0.30/0.35/0.35)
- Integrate: generates standard SKILL.toml + SKILL.md manifests for approved candidates

Thresholds: >=0.7 auto-integrate, 0.4-0.7 manual review, <0.4 skip.
Uses only existing dependencies (reqwest, serde, tokio, tracing, chrono, anyhow).
Includes unit tests for all modules.

* fix: address code review feedback on SkillForge PR #115

- evaluate: whole-word matching for BAD_PATTERNS (fixes hackathon false positive)
- evaluate: guard against future timestamps in recency bonus
- integrate: escape URLs in TOML output via escape_toml()
- integrate: handle control chars (\n, \r, \t, \b, \f) in escape_toml()
- mod: redact github_token in Debug impl to prevent log leakage
- mod: fix auto_integrated count when auto_integrate=false
- mod: per-candidate error handling (single failure no longer aborts pipeline)
- scout: add 30s request timeout, remove unused token field
- deps: enable chrono serde feature for DateTime serialization
- tests: add hackathon/exact-hack tests, update escape_toml test coverage

* fix: address round-2 CodeRabbit review feedback

- integrate: add sanitize_path_component() to prevent directory traversal
- mod: GitHub scout failure now logs warning and continues (no pipeline abort)
- scout: network/parse errors per-query use warn+continue instead of ?
- scout: implement std::str::FromStr for ScoutSource (replaces custom from_str)
- tests: add path sanitization tests (traversal, separators, dot trimming)

---------

Co-authored-by: stawky <stakeswky@gmail.com>
2026-02-15 09:26:13 -05:00
argenis de la rosa
085b57aa30 refactor: consolidate CLI command definitions to lib.rs
- Move all CLI command enums (ChannelCommands, SkillCommands, CronCommands, IntegrationCommands, MigrateCommands, ServiceCommands) to lib.rs
- Add clap derives for use in main.rs CLI parsing
- Update all modules to use crate:: prefix instead of super:: for command types
- Add mod util; to main.rs for binary compilation
- Export Config type from lib.rs for main.rs

This refactoring eliminates code duplication between library modules and binary, centralizing all CLI command definitions in one place.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-15 06:52:33 -05:00
argenis de la rosa
2c7021e90f fix: add memory config to wizard and fix clippy warnings
- Add chunk_max_tokens field to MemoryConfig in quick setup
- Add memory_backend parameter to run_quick_setup()
- Add setup_memory() step to interactive wizard (8 steps now)
- Fix clippy if_not_else warning
- Fix clippy match_same_arms warning
- Add clippy allows for browser.rs (too_many_lines, unnecessary_wraps)
2026-02-14 15:50:53 -05:00
argenis de la rosa
153d6ff149 fix: resolve clippy warnings and formatting issues for CI
- Fix doc_markdown warnings in WhatsApp channel
- Fix needless_pass_by_value in cron, health, migration, service modules
- Fix match_same_arms in migration.rs
- Fix too_many_lines in skills/mod.rs
- Fix manual_let_else in tools/file_write.rs
- Apply cargo fmt formatting fixes

All 435 tests pass, clippy clean.
2026-02-14 15:36:19 -05:00
argenis de la rosa
5476195a7f refactor: remove AIEOS identity support
- Remove src/identity/ directory (aieos.rs, mod.rs)
- Remove IdentityConfig struct and identity field from Config
- Remove build_system_prompt_with_identity and load_aieos_from_config functions
- Remove AIEOS-related imports from channels/mod.rs
- Remove identity module declarations from main.rs and lib.rs
- Remove AIEOS tests from config/schema.rs
- Keep OpenClaw markdown-based identity as the only supported format

This simplifies the codebase by removing unused AIEOS complexity.
All 832 tests pass.
2026-02-14 14:05:14 -05:00
argenis de la rosa
dbf02291b4 fix: escape AppleScript target parameter in iMessage channel
- Add escape_applescript() function to prevent injection attacks
- Add is_valid_imessage_target() validation for phone/email patterns
- Update send() method to escape both message AND target parameters
- Add 40 comprehensive tests covering injection edge cases
- Addresses CWE-78 (OS Command Injection) vulnerability

Fixes #29
2026-02-14 13:38:13 -05:00
argenis de la rosa
ef4444ba43 fix: resolve build errors and add comprehensive symlink tests
- Fixed E0425 error in src/skills/mod.rs by moving println! inside #[cfg(unix)] block where 'dest' variable is in scope
- Added missing 'identity' field to Config struct initializations in src/onboard/wizard.rs
- Fixed import paths for AIEOS identity functions in src/channels/mod.rs
- Added comprehensive symlink edge case tests in src/skills/symlink_tests.rs
- All 840 tests passing, 0 clippy warnings

Resolves issue #28: skills symlink functionality now works correctly on Unix platforms with proper error handling on non-Unix platforms
2026-02-14 13:37:27 -05:00
argenis de la rosa
ec2d5cc93d feat: enhance agent personality, tool guidance, and memory hygiene
- Expand communication style presets (professional, expressive, custom)
- Enrich SOUL.md with human-like tone and emoji-awareness guidance
- Add crash recovery and sub-task scoping guidance to AGENTS.md scaffold
- Add 'Use when / Don't use when' guidance to TOOLS.md and runtime prompts
- Implement memory hygiene system with configurable archiving and retention
- Add MemoryConfig options: hygiene_enabled, archive_after_days, purge_after_days, conversation_retention_days
- Archive old daily memory and session files to archive subdirectories
- Purge old archives and prune stale SQLite conversation rows
- Add comprehensive tests for new features
2026-02-14 11:28:39 -05:00
argenis de la rosa
3d91c40970 refactor: simplify CLI commands and update architecture docs
1. Simplify CLI:
   - Make 'onboard' quick setup default (remove --quick)
   - Add --interactive flag for full wizard
   - Make 'status' detailed by default (remove --verbose)
   - Remove 'tools list/test' and 'integrations list' commands
   - Add 'channel doctor' command
2. Update Docs:
   - Update architecture.svg with Channel allowlists, Browser allowlist, and latest stats
   - Update README.md with new command usage and browser/channel config details
3. Polish:
   - Browser tool integration
   - Channel allowlist logic (empty = deny all)
2026-02-14 05:17:16 -05:00
argenis de la rosa
589921bbf8 feat: add --quick flag to onboard for non-interactive setup
- zeroclaw onboard --quick: generates config with sensible defaults, zero prompts
- zeroclaw onboard --quick --api-key sk-... --provider anthropic: one-liner setup
- Fixes wizard hanging in non-TTY / IDE terminals
- Scaffolds workspace files, prints summary, shows next steps
- 1,017 tests, 0 clippy warnings
2026-02-14 03:19:00 -05:00
argenis de la rosa
c8d4ceee71 feat: add port 0 (random port) support for gateway security
When --port 0 is passed, the OS assigns a random available ephemeral
port (typically 49152-65535). The actual port is resolved after binding
and used for all log output and tunnel forwarding.

This prevents port-scanning attacks against a known fixed port.

Changes:
  src/gateway/mod.rs — bind first, extract actual_port from listener,
    use actual_port for addr formatting and tunnel.start()
  src/main.rs — update CLI help text, conditional log for port=0

8 new edge case tests:
  - port_zero_binds_to_random_port
  - port_zero_assigns_different_ports
  - port_zero_assigns_high_port
  - specific_port_binds_exactly
  - actual_port_matches_addr_format
  - port_zero_listener_accepts_connections
  - duplicate_specific_port_fails
  - tunnel_gets_actual_port_not_zero

943 tests passing, 0 clippy warnings, cargo fmt clean
2026-02-14 01:21:55 -05:00
argenis de la rosa
390cbc0a6c feat: agnostic tunnel system — bring your own tunnel provider
New Tunnel trait + 5 implementations:
- NoneTunnel: local-only, no external exposure (default)
- CloudflareTunnel: wraps cloudflared binary, extracts public URL
- TailscaleTunnel: tailscale serve (tailnet) or funnel (public)
- NgrokTunnel: wraps ngrok binary, supports custom domains
- CustomTunnel: user-provided command with {port}/{host} placeholders

Config schema:
- [tunnel] section with provider selector
- Provider-specific sub-configs: cloudflare, tailscale, ngrok, custom
- Backward compatible (serde default = "none")

Gateway integration:
- Tunnel starts automatically on 'zeroclaw gateway'
- Prints public URL on success, falls back to local on failure

20 new tests (factory, constructors, NoneTunnel async start/health)
649 tests passing, 0 clippy warnings, cargo fmt clean
2026-02-13 16:25:01 -05:00
argenis de la rosa
bc31e4389b style: cargo fmt — fix all formatting for CI
Ran cargo fmt across entire codebase to pass CI's cargo fmt --check.
No logic changes, only whitespace/formatting.
2026-02-13 16:03:50 -05:00
argenis de la rosa
05cb353f7f feat: initial release — ZeroClaw v0.1.0
- 22 AI providers (OpenRouter, Anthropic, OpenAI, Mistral, etc.)
- 7 channels (CLI, Telegram, Discord, Slack, iMessage, Matrix, Webhook)
- 5-step onboarding wizard with Project Context personalization
- OpenClaw-aligned system prompt (SOUL.md, IDENTITY.md, USER.md, AGENTS.md, etc.)
- SQLite memory backend with auto-save
- Skills system with on-demand loading
- Security: autonomy levels, command allowlists, cost limits
- 532 tests passing, 0 clippy warnings
2026-02-13 12:19:14 -05:00