Commit Graph

16 Commits

Author SHA1 Message Date
Levi Neely b4baddd826 Memory tools: never expose raw memo commands to agents
- Add MEMO_TOOLS=1 env var to memo script; when set, all printed
  instructions reference native tool names instead of memo paths
- Set MEMO_TOOLS=1 in all memory tool .meta wrappers
- Add memory_nap tool for compressions
- Add memory_zoom tool for tree navigation
- Add part/T pagination args to memory_wake
- Update system prompt to use memory_zoom tool call
- Fix inject ctl: submit as user message when agent is idle
2026-08-19 10:13:02 +02:00
Ollie Agent 94b4c4be3f Integrate OptMem wake into agent memory prompting 2026-08-18 14:01:18 +02:00
Levi Neely 74ba5ae7ac remove maxSteps and ResetsCounter entirely
The step budget mechanism is gone. Agents run until they finish,
are interrupted by the user, or (for sub-agents) hit the timeout.

No replacement. The human is the kill switch.
2026-08-17 09:48:58 +02:00
Ollie Agent f82b36ea80 Integrate OptMem memory backend 2026-08-16 11:10:50 +02:00
Ollie Agent 50e465336e Add Tree-sitter code intelligence tools 2026-08-15 21:13:40 +02:00
Levi Neely eca26bd6d4 default agent: add subagent_spawn to autoLoad 2026-08-14 14:56:51 +02:00
Levi Neely 16e54d1f03 fix: trailing comma in agent configs (invalid JSON from sed) 2026-08-13 21:49:58 +02:00
Levi Neely a1e78bca80 remove tool_load: use ctl directly
tool_load was a built-in intercept in the agent loop — the only
'tool' that didn't run in toolsrv. Removed entirely:
- Intercept in loop.go (25 lines)
- Script + .meta in data/tools/
- autoLoad references in agent configs

Loading tools is now exclusively via ctl (which already existed):
  echo 'tool_load X' | ollie-9p write .../ctl

System prompt updated to show the ctl pattern.
2026-08-13 21:44:46 +02:00
Ollie Agent 2d74ffb95e prompt audit: trim system prompt, kill dead code, improve maintainability
- System prompt: ~200 → ~80 lines (removed tool-first table, API docs,
  output protocol, dead 9P entries)
- Removed AllowTools entirely (field, RPC, configs)
- Removed PRIME_* env vars, replaced with typed struct fields
- Merged tool listing + docs into single renderTools()
- Killed BuildToolListing, changed OnToolsChanged to func() signal
- Typed Preamble.Section (compile-time safety)
- Added protocol docs on extractToolResult
- Added session/ package godoc
- Moved output protocol to per-agent prompts
- Rewrote data/agents/README.md
- Moved 7 reference docs from doc/resources/ to doc/
- Deleted PromptEnv() dead function
2026-08-09 14:13:52 +02:00
Levi Neely e84b4ffef7 feat: add resetsCounter tool property to reset step budget on active progress
Tools like file_write, file_edit, and shell now reset the agent's step
counter when called successfully. This allows the agent to continue
working without hitting the soft step-budget guardrail as long as it's
making active progress (writing files, running commands) rather than
looping on research/reading.

Changes:
- Add ResetsCounter bool to ToolInfo and MetaFile structs
- Parse resetsCounter from .meta JSON files
- Reset step counter in agent loop when ResetsCounter tool succeeds
- Add resetsCounter: true to file_write, file_edit, shell
- Reduce maxSteps from 50 to 25 (resetsCounter makes this safe)
2026-08-07 17:13:14 +02:00
Ollie Agent 06b287b8f2 remove hooks blocks and beads/freeloader refs from all agent JSON configs
- Remove "hooks" blocks from all 8 agent JSON files (copilot, default, driver, explorer, librarian, navigator, taskmanager, theo)
- Remove beads shell one-liner from prompt arrays in default.json and driver.json
- Hooks mechanism is dead code; these empty blocks are just noise
2026-08-07 07:15:04 +02:00
Ollie Agent d80a87eb86 clean up OLLIE_* env vars: replace with XDG conventions
- Remove all OLLIE_*_PATH vars (TOOLS_PATH, CFG_PATH, DATA_PATH, etc.)
  Use XDG_CONFIG_HOME/ollie/* and XDG_DATA_HOME/ollie/* instead
- Replace OLLIE_<TAG>_LOG per-component logging with single OLLIE_LOG={level}
- Replace OLLIE_USAGE_LOG with XDG_DATA_HOME-based path
- Replace OLLIE_OLLAMA_URL with standard OLLAMA_HOST (or omit entirely)
- Rename protocol markers: OLLIE_LISTEN_READY → ListenReady, OLLIE_9P_OPEN → Open
- Remove freeloader hooks from agent configs
- Update sandbox YAMLs to use XDG paths instead of OLLIE_*_PATH tokens
- Update shell tools to use $(dirname "$0") fallback instead of OLLIE_TOOLS_PATH
- Update docs and prompts accordingly
2026-08-07 07:11:53 +02:00
Ollie Agent 7beef106e0 tool_load: add dynamic tool loader with bash script
Add a tool_load tool that writes a tool name to the session's 9P
tools file via ollie-9p, enabling on-demand tool loading.

Reduce autoLoad lists in default.json and theo.json to bare
essentials — everything else is loaded via tool_load.
2026-08-03 00:15:36 +02:00
Ollie Agent 088aa38f1c data/agents/default.json: remove hard-coded shell and reasoning_think from autoLoad
These two are built-in tools always available regardless of autoload.
No need to explicitly load them.
2026-08-02 21:16:43 +02:00
Ollie Agent e25157ffcd multiplex toolsrv, wip zero-tool server 2026-08-02 19:18:39 +02:00
Ollie Agent 942cad2dee rename contrib/{prompts,scripts,services,tools,skills,agents} → data/
These are project data files, not community contributions.
contrib/elisp stays as-is.
2026-07-30 21:40:11 +02:00