Add semantic skill matching using all-MiniLM-L6-v2 sentence embeddings.
Skills are automatically injected into user turns based on relevance.
New packages:
- embedding: ONNX-based text embedding with MiniLM model
- skills: skill discovery, embedding cache, and semantic matching
Integration:
- InitSkillIndex called at startup in fs.NewRoot
- matchSkills called per-turn in executeTurn
- Matched skills injected in <context> block alongside user prompts
Makefile:
- install-models target downloads model and ONNX runtime
- Model files stored in ~/.local/share/ollie/models/
Config:
- Threshold: 0.2 cosine similarity
- Limit: 3 skills per turn
- Skill dirs: ~/.kiro/skills (user), ~/.config/ollie/skills (installed)
- Add quirks package for stupid model behavior workarounds
- ShellInvokesNativeTool blocks shell(cmd="tool_name") patterns
- Add client_9p tool: native wrapper for ollie-9p operations
- Block ollie-9p in shell — use client_9p instead
- Update all prompts to use client_9p, not shell+ollie-9p
- Clarify 9P namespace is complete (tools are NOT in 9P)
- Registry.All() lists all available tools for validation
Add explicit guidance to Autonomous Operation section:
- New unacceptable behaviors: giving up before checking skill_list or trying to load tools
- New checklist: must check skills, try loading tools, attempt even uncertain options before reporting failure
When an agent calls a tool that exists but isn't loaded, toolsrv now
automatically loads it instead of returning an error. This eliminates
the round-trip of a separate load call.
- Remove Registry.Exists() (superseded by auto-load logic)
- Fire OnToolsChanged callback after auto-load so olliesrv can update
tool definitions for subsequent turns
chat.raw includes everything; chat strips:
- [[[...]]] markers
- fence delimiters
- <context>...</context> blocks (user prompts)
Text clients get clean output; GUI clients use chat.raw for full control.
Reduced from 18 targets to 10:
all, build, go, tools, kde, install, install-data, test, clean, uninstall
Key changes:
- Build to build/ dir first, install only after tests pass
- Merged ninep+core into 'go' (builds olliesrv, ollie-9p, toolsrv)
- Removed build-kf5/install-kf5/kde-kf5 (cmake auto-detects)
- Merged test-core+test-9p into 'test'
- Removed 'client' target (lib9p now built by cmake as custom command)
- Added 'uninstall' target
CMakeLists.txt changes:
- lib9p built via add_custom_command instead of requiring pre-built .so
- add_dependencies ensures lib9p built before KDE targets need it
user-preferences.md is meant to be prepended to every user message,
not part of the system prompt. Moved from 'prompt' to 'userPrompts'
in: author, conductor, default, foreman, panelist, reviewer.
Other agents already had it in the correct location.
Added explicit unacceptable behaviors:
- Mentioning a tool without calling it
- Acknowledging a tool exists but not using it
Added bold statement: 'Knowing a tool exists is not the same as using it.
Your response should contain tool calls, not descriptions of tools you
could call.'
Agents were using shell to invoke tools (cat, grep, ollie-9p) instead of
calling the actual tools. Added unmistakable WRONG/RIGHT code blocks
showing the correct pattern. Shell is ONLY for git, make, npm, etc.
Rewrote system prompt to enforce immediate tool use:
- New 'Autonomous Operation' section: act first, report results
- Explicit list of unacceptable behaviors (narrating intentions, asking
permission for routine ops, producing text when tools should be called)
- Tools section: 'Use them without hesitation', concrete examples
- Skills section: 'if the task needs it, load it' — no asking
- Stronger sub-agent prefix: 'Do NOT respond with a plan. Call tools.'
Timeout fix:
- Add Timeout field to ToolInfo (protocol) and MetaFile (metadata)
- proc.go respects tool-declared timeout before falling back to 30s default
- subagent_spawn.meta declares timeout=0 (no timeout) so the tool is
never killed prematurely while waiting for the sub-agent to finish
- Tool schema declares timeout with 'do not set' guidance to prevent
the LLM from adding a short timeout
Premature response fix:
- Inject behavioral prefix into sub-agent prompt: complete all work
before responding, report results not intentions
- Sub-agent's final text is returned to parent; this instruction ensures
it contains accomplished work, not a plan
QIcon::fromTheme("ollie") was finding a stale brain SVG at
/usr/share/icons/ instead of the octopus. Now the app uses the
embedded resource for both window icon and systray, and sets
app.setWindowIcon() explicitly.
Also fixed the .desktop file to use the theme name 'ollie' instead
of a hardcoded absolute path.
killTargetSessionName was set to item.name (displayName, e.g. 'ollie (1)')
instead of item.baseName (the actual 9P path component 'ollie').
This caused readSessionConfig() to read from a non-existent path.
Also fixes the Goal dialog which had the same latent bug.
- Workflow() no longer defaults to 'conductor' when empty
- runWorkflow() returns immediately for '' or 'none'
- 'none' listed first in the workflows file output
- QML dropdowns default to 'none' instead of 'conductor'
- C++ fallback uses 'none' when server unreachable
- Add session-level cfg file (read: name/cwd/remote/workflow/variant/yolo;
write: workflow, variant, cwd)
- Add readSessionConfig() and updateSessionConfig() to C++ client
- NewSessionDialog supports editMode: pre-populates fields from session cfg,
title becomes 'Session Settings', button becomes 'Save'
- Name and Remote fields disabled in edit mode (non-reconfigurable)
- YOLO checkbox hidden in edit mode
- Only changed values are written back on save
- Add 'Settings...' to session context menu in SessionTree
- Wire sessionSettingsRequested signal through to dialog
- loadWorkflows() parses tab-separated 'name<TAB>variants' format
- Add variantsForWorkflow() to return variant list for a workflow
- Add m_workflowVariants map to cache per-workflow variant lists
- createSession() accepts variant parameter, passes to server
- QML: separate Workflow and Variant ComboBoxes
- Variant dropdown repopulates on workflow selection change
- 'default' always present; non-default variants pass through to launcher
- runWorkflow accepts a variant parameter; sources {workflow}-{variant}.conf
as env vars before exec'ing the script
- Session stores variant; persisted and restored
- session/new accepts variant= parameter
- 'run' ctl command accepts optional variant as second arg
- workflows file now lists variants: name<TAB>default,variant1,...
- review workflow reads AUTHOR_PROFILE/REVIEWER_PROFILE env vars
- Add review-code.conf and review-writing.conf example variants
Add peer/ directory to each agent's 9P namespace. Agents communicate
by writing to peer/{name}, which delivers to the target's prompt handler.
Only declared peers can be messaged — the directory is the ACL.
Implementation:
- Agent struct: peers map + AddPeer/RemovePeer/Peers methods
- fs/spec.go: peer/ Each node (write-only entries), peeradd/peerdel/peers ctl commands
- Bidirectional: peeradd A on B also adds B on A
- Peers constrained to same session
- Persisted with session state (PersistedAgent.Peers field)
- peeradd/peerdel trigger immediate session save
Docs updated: system_prompt.md, AGENTS.md, README.md, architecture-9p.md,
architecture-core.md, architecture.md, usage.md.
All tool logic lives in the multicall main.go at each family root.
Move .meta files to data/tools/ (single canonical location).
Expand test-core to cover all tools.
Moves package lsp and cmdutil under tools/lsp/internal/ so the
multicall binary can live at tools/lsp/main.go — same layout as
codeintel and filetools.
- Multicall pattern for tool families: one binary per family with
symlinks for each tool name. Reduces tools dir from 156MB to 31MB.
codeintel (6 tools), filetools (5 tools), lsptools (7 tools).
- Build with -ldflags="-s -w" to strip debug symbols.
- Fix Landlock EINVAL on regular files: filter directory-only access
rights (READ_DIR, MAKE_*, REMOVE_*) when adding rules for non-
directory paths.
- Fix toolsrv CWD initialization: pass --cwd flag value to server
state at startup so {CWD} in sandbox.yaml expands correctly.
- 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
- Add OptMem memory section to system_prompt.md with tool-based API
- Fix KRunner plugin icon: resolve via QStandardPaths instead of theme name
- Fix desktop file icon: use absolute path to bypass stale system icon
Ollie Agent
Levi Neely