m3tam3re/AGENTS
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
74 Commits 1 Branch 0 Tags
a88e6f9f069cbee23f638bcbe6831cd94b6d0dec
Commit Graph

74 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
m3tm3re
a88e6f9f06 docs(basecamp-project): add formatting rules reference 2026-04-27 09:46:20 +02:00
m3tm3re
6e0e847299 feat: basecamp-project skill 2026-04-24 20:00:33 +02:00
m3tm3re
0ad41acb03 merge 2026-04-13 17:05:21 +02:00
m3tm3re
ccca3dd9db feat: config with agents rework 2026-04-13 16:53:17 +02:00
m3tm3re
a81e178856 feat: export loadAgents and backward-compat agentsJson from flake 2026-04-10 16:31:33 +02:00
m3tm3re
7a8dd525c9 feat: add canonical agent.toml definitions for all 6 agents 2026-04-10 16:03:58 +02:00
m3tm3re
cd553b4031 docs: add canonical agent.toml schema definition 2026-04-10 15:59:29 +02:00
sascha.koenig
a05558b811 chore: repo cleanup 2026-03-31 19:13:10 +02:00
sascha.koenig
586d1484ec feat: external skills 2026-03-27 14:23:55 +01:00
sascha.koenig
a4ae041e1d +kestra skills 2026-03-16 10:47:57 +01:00
m3tm3re
8bcb5e11fb feat(skills): add qmd skill for vault search and memory, deprecate SQLite memory 
- Add skills/qmd/SKILL.md: qmd-based knowledge retrieval and memory system
  - Search workflows: BM25, vector semantic, hybrid+rerank modes
  - Memory write patterns via direct filesystem writes to ~/CODEX/80-memory/
  - Session summary workflow for capturing work at session end
  - Auto-recall pattern for context injection at session start

- Deprecate skills/memory/SKILL.md: SQLite-based memory replaced by qmd
  - qmd provides superior hybrid search over entire vault
  - No external OpenAI embedding dependency
  - Uses local GPU (7900 XTX) for embeddings and reranking

- Update skills/obsidian/SKILL.md: remove memory search sections
  - Memory search now handled by qmd skill
  - Keep CRUD operations and daily notes functionality
  - Update integration table to reference qmd

Related: CODEX vault updated with memory template, session-summary template,
and sessions directory. qmd collections and context annotations configured.
 2026-03-14 10:04:43 +01:00
m3tm3re
39ac89f388 docs: update AGENTS.md and README.md for rules system, remove beads 
- Add rules/ directory documentation to both files
- Update skill count from 25 to 15 modules
- Remove beads references (issue tracking removed)
- Update skills list with current active skills
- Document flake.nix as proper Nix flake (not flake=false)
- Add rules system integration section
- Clean up sisyphus planning artifacts
- Remove deprecated skills (memory, msteams, outlook)
 2026-03-03 19:40:57 +01:00
m3tm3re
1bc81fb38c chore: update readme 2026-02-18 17:32:13 +01:00
m3tm3re
1f1eabd1ed feat(rules): add strict TDD enforcement ruleset with AI patterns 2026-02-18 17:30:20 +01:00
m3tm3re
5b204c95e4 test(rules): add final QA evidence and mark review complete 
Final Review Results:
- F1 (Plan Compliance): OKAY - Must Have [12/12], Must NOT Have [8/8]
- F2 (Code Quality): OKAY - All files pass quality criteria
- F3 (Manual QA): OKAY - Scenarios [5/5 pass]
- F4 (Scope Fidelity): OKAY - No unaccounted changes

All 21 tasks complete (T1-T17 + F1-F4)
 2026-02-17 19:31:24 +01:00
m3tm3re
4e9da366e4 test(rules): add integration test evidence 
- All 11 rule files verified (exist, under limits)
- Full lib integration verified (11 paths returned)
- Context budget verified (975 < 1500)
- All instruction paths resolve to real files
- opencode.nix rules entry verified

Refs: T17 of rules-system plan
 2026-02-17 19:18:39 +01:00
m3tm3re
8910413315 feat(rules): add initial rule files for concerns, languages, and frameworks 
Concerns (6 files):
- coding-style.md (163 lines): patterns, anti-patterns, error handling, SOLID
- naming.md (105 lines): naming conventions table per language
- documentation.md (149 lines): docstrings, WHY vs WHAT, README standards
- testing.md (134 lines): AAA pattern, mocking philosophy, TDD
- git-workflow.md (118 lines): conventional commits, branch naming, PR format
- project-structure.md (82 lines): directory layout, entry points, config placement

Languages (4 files):
- python.md (224 lines): uv, ruff, pyright, pytest, pydantic, idioms, anti-patterns
- typescript.md (150 lines): strict mode, discriminated unions, satisfies, as const
- nix.md (129 lines): flake structure, module patterns, alejandra, anti-patterns
- shell.md (100 lines): set -euo pipefail, shellcheck, quoting, POSIX

Frameworks (1 file):
- n8n.md (42 lines): workflow design, node patterns, Error Trigger, security

Context budget: 975 lines (concerns + python) < 1500 limit

Refs: T6-T16 of rules-system plan
 2026-02-17 19:05:45 +01:00
m3tm3re
d475dde398 feat(rules): add rules directory structure and usage documentation 
- Create rules/{concerns,languages,frameworks}/ directory structure
- Add USAGE.md with flake.nix integration examples
- Add plan and notepad files for rules-system implementation

Refs: T1, T5 of rules-system plan
 2026-02-17 18:59:43 +01:00
m3tm3re
6fceea7460 refactor: modernize agent configs, remove beads, update README 
- Upgrade all agents from glm-4.7 to glm-5 with descriptive names
- Add comprehensive permission configs (bash, edit, external_directory) for all agents
- Remove .beads/ issue tracking directory
- Update README: fix opencode URL to opencode.ai, remove beads sections, formatting cleanup
 2026-02-17 09:15:15 +01:00
m3tm3re
923e2f1eaa chore(plan): mark deployment verification as blocked (requires user action) 2026-02-14 08:34:06 +01:00
m3tm3re
231b9f2e0b chore(plan): mark tasks 11-14 and definition of done as complete 2026-02-14 08:31:32 +01:00
m3tm3re
c64d71f438 docs(memory): update skills for opencode-memory plugin, deprecate mem0 2026-02-14 08:22:59 +01:00
m3tm3re
1719f70452 feat(memory): add core memory skill, update Apollo prompt and Obsidian skill 
- Add skills/memory/SKILL.md: dual-layer memory orchestration
- Update prompts/apollo.txt: add memory management responsibilities
- Update skills/obsidian/SKILL.md: add memory folder conventions
 2026-02-12 20:02:51 +01:00
m3tm3re
0d6ff423be Add Memory System configuration to user profile 2026-02-12 19:54:54 +01:00
m3tm3re
79e6adb362 feat(mem0-memory): add memory categories and dual-layer sync patterns 2026-02-12 19:50:39 +01:00
m3tm3re
1e03c165e7 docs: Add Obsidian MCP server configuration documentation 
- Create mcp-config.md in skills/memory/references/
- Document cyanheads/obsidian-mcp-server setup for Opencode
- Include environment variables, Nix config, and troubleshooting
- Reference for Task 4 of memory-system plan
 2026-02-12 19:44:03 +01:00
m3tm3re
94b89da533 finalize doc-translator skill 2026-02-11 19:58:06 +01:00
sascha.koenig
b9d535b926 fix: use POST method for Outline signed URL upload 
Change HTTP method from PUT to POST on line 77 for signed URL upload,
as Outline's S3 bucket only accepts POST requests.
 2026-02-11 14:16:02 +01:00
sascha.koenig
46b9c0e4e3 fix: list_outline_collections.sh - correct jq parsing to output valid JSON array 2026-02-11 14:14:55 +01:00
m3tm3re
eab0a94650 doc-translator fix 2026-02-10 20:24:13 +01:00
m3tm3re
0ad1037c71 doc-translator 2026-02-10 20:02:30 +01:00
m3tm3re
1b4e8322d6 doc-translator 2026-02-10 20:00:42 +01:00
m3tm3re
7a3b72d5d4 chore: mark chiron-agent-framework plan as complete 
All 27 tasks completed successfully.

Co-authored-by: Atlas orchestrator <atlas@opencode.dev>
 2026-02-03 20:40:06 +01:00
m3tm3re
156ebf7d63 docs: fix duplicate success criteria in chiron-agent-framework plan 
All 6 success criteria now properly marked as complete.

Co-authored-by: Atlas orchestrator <atlas@opencode.dev>
 2026-02-03 20:39:26 +01:00
m3tm3re
a57e302727 docs: complete all success criteria in chiron-agent-framework 
All 6 success criteria now marked as complete.

Co-authored-by: Atlas orchestrator <atlas@opencode.dev>
 2026-02-03 20:34:28 +01:00
m3tm3re
d08deaf9d2 docs: mark all success criteria as complete 
All 6 success criteria in plan file now marked complete.

Co-authored-by: Atlas orchestrator <atlas@opencode.dev>
 2026-02-03 20:34:18 +01:00
m3tm3re
666696b17c docs: mark chiron-agent-framework plan complete 
All 14 tasks completed and verified.

## Summary
- 6 agents defined (2 primary, 4 subagents)
- 6 system prompts created
- 5 tool integration skills created
- 1 validation script created
- All success criteria met

Co-authored-by: Atlas orchestrator <atlas@opencode.dev>
 2026-02-03 20:33:01 +01:00
m3tm3re
1e7decc84a feat: add Chiron agent framework with 6 agents and 5 integration skills 
Complete implementation of personal productivity agent framework for Oh-My-Opencode.

## Components Added

### Agents (6 total)
- Primary agents: chiron (Plan Mode), chiron-forge (Build Mode)
- Subagents: hermes (work communication), athena (work knowledge), apollo (private knowledge), calliope (writing)

### System Prompts (6 total)
- prompts/chiron.txt - Main orchestrator with delegation logic
- prompts/chiron-forge.txt - Execution/build counterpart
- prompts/hermes.txt - Basecamp, Outlook, MS Teams specialist
- prompts/athena.txt - Outline wiki/documentation specialist
- prompts/apollo.txt - Obsidian vault/private notes specialist
- prompts/calliope.txt - Writing/documentation specialist

### Integration Skills (5 total)
- skills/basecamp/SKILL.md - 63 MCP tools documented
- skills/outline/SKILL.md - Wiki/document management
- skills/msteams/SKILL.md - Teams/channels/meetings
- skills/outlook/SKILL.md - Email/calendar/contacts
- skills/obsidian/SKILL.md - Vault/note management

### Validation
- scripts/validate-agents.sh - Agent configuration validation
- All agents validated: JSON structure, modes, prompt references
- All prompts verified: Exist, non-empty, >500 chars
- All skills verified: Valid YAML frontmatter, SKILL.md structure

## Verification
✅ 6 agents in agents.json
✅ All 6 prompt files exist and non-empty
✅ All 5 skills have valid SKILL.md with YAML frontmatter
✅ validate-agents.sh passes (exit 0)

Co-authored-by: Sisyphus framework <atlas@opencode.dev>
 2026-02-03 20:30:34 +01:00
m3tm3re
76cd0e4ee6 Create Athena (Work Knowledge) system prompt 
- Outline wiki specialization: document CRUD, search, collections, sharing
- Focus: wiki search, knowledge retrieval, documentation updates
- Follows standard prompt structure: 8 sections matching Apollo/Calliope
- Explicit boundaries: Hermes (comm), Apollo (private), Calliope (creative)
- Uses Question tool for document selection and search scope
- Verification: outline, wiki/knowledge, document keywords confirmed
 2026-02-03 20:18:52 +01:00
m3tm3re
4fcab26c16 Create Hermes system prompt (Wave 2, Task 5) 
- Added prompts/hermes.txt with Basecamp, Outlook, Teams specialization
- Follows consistent structure pattern from apollo.txt and calliope.txt
- Defines Hermes as work communication specialist
- Includes tool usage patterns for Question tool and MCP integrations
- Verifies with grep: basecamp, outlook/email, teams/meeting
- Appends learnings to chiron-agent-framework notepad
 2026-02-03 20:18:46 +01:00
m3tm3re
f20f5223d5 Create agents.json with 6 agent definitions (Wave 1, Task 1) 
- Added all 6 agents: chiron, chiron-forge, hermes, athena, apollo, calliope
- Primary agents (2): chiron (Plan Mode), chiron-forge (Build Mode)
- Subagents (4): hermes (communications), athena (work knowledge), apollo (private knowledge), calliope (writing)
- All agents use model: zai-coding-plan/glm-4.7
- Prompt references use file pattern: {file:./prompts/<name>.txt}
- Permission structure: primaries have external_directory rules, subagents have simple question: allow
- Verified with Python JSON validation (6 agents, correct names)
- Documented patterns and learnings in notepad
 2026-02-03 20:14:34 +01:00
m3tm3re
36c82293f9 Agent restructure 2026-02-03 20:09:15 +01:00
m3tm3re
7e4a44eed6 Agent restructure 2026-02-03 20:04:26 +01:00
m3tm3re
1f320f1c95 Add scripts/validate-agents.sh for agent validation 2026-02-03 19:23:26 +01:00
m3tm3re
fddc22e55e Add outlook skill with Graph API documentation 
- Create skills/outlook/SKILL.md with comprehensive Outlook Graph API documentation
- Document mail CRUD operations: list, get, create, send, reply, forward, update, delete
- Document folder management: list, create, update, delete, move, copy
- Document calendar events: list, get, create, update, delete, accept/decline
- Document contacts: list, get, create, update, delete, folder management
- Include search operations for mail, contacts, and events
- Provide common workflows for email, inbox organization, meeting invitations
- Include IDs and discovery guidance
- Set compatibility to opencode
- Close issue AGENTS-ch2
 2026-02-03 18:55:15 +01:00
m3tm3re
db1a5ba9ce Add MS Teams Graph API integration skill 
Created skills/msteams/SKILL.md with comprehensive documentation for:
- Teams and channels management
- Channel messages (send, retrieve, edit, delete)
- Meeting scheduling and management
- Chat conversations (1:1, group, meeting)
- Common workflows for automation
- API endpoint reference
- Best practices and integration examples

Follows SKILL.md format with YAML frontmatter.
Compatibility: opencode
 2026-02-03 18:52:14 +01:00
m3tm3re
730e33b908 Add Apollo system prompt for private knowledge management 2026-02-03 18:50:32 +01:00
m3tm3re
ecece88fba Create Calliope writing prompt 
- Define Calliope as Greek muse specializing in documentation, reports, meeting notes
- Include Question tool for clarifying tone, audience, format
- Set scope boundaries: delegates tools, no overlap with Hermes/Athena
- Follow standard prompt structure from agent-development skill
 2026-02-03 18:50:22 +01:00
m3tm3re
1252b9ffe7 Create Chiron-Forge build/execution mode system prompt 
Define Chiron-Forge as execution/build counterpart to Chiron with:
- Full write access for task execution
- Clear distinction from Chiron's planning/analysis role
- Question tool for destructive operations confirmation
- Workflow: Receive → Understand → Plan Action → Execute → Confirm → Report
- Delegation to subagents for specialized domains

File: prompts/chiron-forge.txt (3185 chars, 67 lines)
 2026-02-03 18:49:31 +01:00
m3tm3re
3f2f766af6 Create prompts/chiron.txt with Chiron plan/analysis mode system prompt 
- Define Chiron as main orchestrator in plan/analysis mode
- Include delegation logic to subagents (Hermes, Athena, Apollo, Calliope)
- Add Question tool usage for ambiguous requests
- Specify read-only permissions (no file modifications)
- Focus on planning, analysis, guidance, and delegation
- Use second-person addressing throughout
 2026-02-03 18:48:39 +01:00