Files

zhi 0729e83b38 feat: split dirigent_tools into individual tools + human @mention override

Feature 1: Split dirigent_tools
- Replace monolithic dirigent_tools (9 actions) with 9 individual tools
- Discord: dirigent_channel_create, dirigent_channel_update, dirigent_member_list
- Policy: dirigent_policy_get, dirigent_policy_set, dirigent_policy_delete
- Turn: dirigent_turn_status, dirigent_turn_advance, dirigent_turn_reset
- Extract shared executeDiscordAction() helper

Feature 2: Human @mention override
- When humanList user @mentions agents, temporarily override turn order
- Only mentioned agents cycle, ordered by their turn order position
- Original order restores when cycle returns to first agent or all NO_REPLY
- New: setMentionOverride(), hasMentionOverride(), extractMentionedUserIds()
- New: buildUserIdToAccountIdMap() for reverse userId→accountId resolution

Bump version to 0.3.0

2026-03-07 16:55:01 +00:00

2.4 KiB

Raw Blame History

Changelog

0.3.0

Split dirigent_tools into individual tools: Each action is now a separate tool with focused parameters:
- dirigent_channel_create, dirigent_channel_update, dirigent_member_list (Discord control)
- dirigent_policy_get, dirigent_policy_set, dirigent_policy_delete (policy management)
- dirigent_turn_status, dirigent_turn_advance, dirigent_turn_reset (turn management)
Human @mention override: When a humanList user @mentions specific agents:
- Temporarily overrides the speaking order to only mentioned agents
- Agents cycle in their original turn-order position
- After all mentioned agents have spoken, original order restores and state goes dormant
- Handles edge cases: all NO_REPLY, reset, non-agent mentions

0.2.0

Project renamed from WhisperGate to Dirigent
- All plugin ids, tool names, config keys, file paths, docs updated
- Legacy whispergate config key still supported as fallback
Identity prompt enhancements: Discord userId now included in agent identity injection
Scheduling identifier: Added configurable schedulingIdentifier (default: ➡️)
- Moderator handoff now sends <@USER_ID>➡️ instead of semantic messages
- Agent prompt explains the identifier is meaningless — check chat history and decide
All prompts in English: End-marker instructions, group chat rules, slash command help text

0.1.0-mvp

Added no-reply API service (/v1/chat/completions, /v1/responses, /v1/models)
Added optional bearer auth (AUTH_TOKEN)
Added plugin with deterministic rule gate
Added discord-specific 🔚 prompt injection for bypass/end-symbol paths
Added containerization (Dockerfile, docker-compose.yml)
Added helper scripts for smoke/dev lifecycle and rule validation
Added no-touch config rendering and integration docs
Added installer script with rollback (scripts/install-dirigent-openclaw.sh)
- supports --install / --uninstall
- uninstall restores all recorded changes
- writes install/uninstall records under ~/.openclaw/dirigent-install-records/
Added discord-control-api with:
- channel-private-create (create private channel for allowlist)
- channel-private-update (update allowlist/overwrites for existing channel)
- member-list (guild members list with pagination + optional field projection)
- guardrails: action mode validation, id-list limits, response-size limit

2.4 KiB Raw Blame History

Changelog

0.3.0

0.2.0

0.1.0-mvp

2.4 KiB

Raw Blame History