niri: drop untracked noctalia.kdl include; sync memory skills
- remove 'include noctalia.kdl' from shared config.kdl: the file is a per-host noctalia template render (colors from local palette), was never committed, and made every config load fail hard on hosts where it was absent — niri 26.04 falls back to DEFAULT config on parse error, silently killing keybinds/rules/layout/autostart - theme integration now only via keybinds (noctalia msg verbs) + the host-local rendered files (see managed skill noctalia-template-offline-render) - sync agent memories: delete retired memory-store skills, add new ones
This commit is contained in:
Frank committed
1 parent
fc049ce7fb
commit
970ceed554
33 files changed
+200
-62
No files matched your search
@@ -1,5 +0,0 @@
|
||||
Handle FastFlowLM (FLM) image constraints for NPU inference.
|
||||
- Endpoint: `http://192.168.0.143:7771/v1`
|
||||
- Strict decoder: Accepts only PNG and JPEG/JPG. WebP is rejected.
|
||||
- Workflow: Convert WebP upstream before FLM calls.
|
||||
- Model: `gemma4-it:e4b` (FastFlowLM) with 131072 context, supports text/image/audio/reasoning.
|
||||
@@ -1,5 +0,0 @@
|
||||
#!/bin/bash
|
||||
# Convert WebP to PNG for FLM NPU endpoint
|
||||
for file in *.webp; do
|
||||
magick "$file" "${file%.webp}.png"
|
||||
done
|
||||
@@ -0,0 +1 @@
|
||||
Standard Git workflow for `~/.omp`. Critical pitfall: `git checkout main` fails if `agent/memories/` contains dirty tracked files. Always run `git stash` before checkout. Stash saves on the source branch. PR creation uses `pr/fix-<feature>` branches. Gitea at `dankclaw:3000`.
|
||||
+5
@@ -0,0 +1,5 @@
|
||||
Workflow for interrupted config edits:
|
||||
1. Edit files in `develop`
|
||||
2. `git stash` before switching to `main`
|
||||
3. Commit on `develop`, apply stash later
|
||||
4. Resolves dirty-file checkout block cleanly.
|
||||
@@ -0,0 +1,9 @@
|
||||
#!/bin/bash
|
||||
# Safe checkout for ~/.omp with memory directory awareness
|
||||
set -euo pipefail
|
||||
if ! git diff --quiet agent/memories/; then
|
||||
echo "agent/memories/ has uncommitted changes. Stashing..."
|
||||
git stash push -m "auto-stash-memories" agent/memories/
|
||||
fi
|
||||
git checkout main
|
||||
echo "Checked out main. Stash available on develop if needed."
|
||||
@@ -0,0 +1,5 @@
|
||||
## PR: <feature-fix>
|
||||
- Branch: `pr/fix-<feature>`
|
||||
- Base: `main`
|
||||
- Scope: Single clean commit, focused change.
|
||||
- Notes: Verify `agent/memories/` stashed if dirty.
|
||||
@@ -1,5 +0,0 @@
|
||||
Dual-edit model configuration between `~/.pi/agent/models.json` and `~/.omp/agent/models.yml`.
|
||||
- Use Ki-style multiples for context (e.g., 200k → 204800).
|
||||
- Update both `contextWindow` and `maxTokens` simultaneously.
|
||||
- No cache flush required for config-only updates.
|
||||
- Validation: `jq -e '.providers."llamacpp (dankserver)".models[0] | .contextWindow, .maxTokens' ~/.pi/agent/models.json`
|
||||
@@ -1,3 +0,0 @@
|
||||
#!/bin/bash
|
||||
jq -e '.providers."llamacpp (dankserver)".models[0] | .contextWindow, .maxTokens' ~/.pi/agent/models.json
|
||||
python3 -c "import yaml; d=yaml.safe_load(open('/home/dank/.omp/agent/models.yml')); m=d['providers']['llamacpp (dankserver)']['models'][0]; print(m['id'], m['contextWindow'], m['maxTokens'])"
|
||||
@@ -0,0 +1 @@
|
||||
Manages dual AI model configs (`~/.pi/agent/models.json` as source, `~/.omp/agent/models.yml` as import/generated). Always edit both in lock-step. Context sizes use Ki-style multiples (e.g., 200k = 204800). For `frank-default-small`, update `contextWindow` and `maxTokens` together. Validate with `jq` and `python3 -c "import yaml..."`. Providers: `llamacpp (dankserver)` & `llamacpp (dankclaw)`. NPU `gemma4-it:e4b` uses port 7771.
|
||||
+5
@@ -0,0 +1,5 @@
|
||||
When updating context from 160k to 200k:
|
||||
1. Convert to Ki: 200 × 1024 = 204800
|
||||
2. Update `contextWindow` and `maxTokens` in both `~/.pi/agent/models.json` and `~/.omp/agent/models.yml`
|
||||
3. No cache flush required; applies on next config read.
|
||||
4. Validate with validation script.
|
||||
+9
@@ -0,0 +1,9 @@
|
||||
#!/bin/bash
|
||||
# Validate both JSON and YAML context/tokens match
|
||||
JSON_VAL=$(jq -e '.providers."llamacpp (dankserver)".models[0] | .contextWindow, .maxTokens' ~/.pi/agent/models.json)
|
||||
YAML_VAL=$(python3 -c "import yaml; d=yaml.safe_load(open('/home/dank/.omp/agent/models.yml')); m=d['providers']['llamacpp (dankserver)']['models'][0]; print(m['id'], m['contextWindow'], m['maxTokens'])")
|
||||
if [ "$JSON_VAL" = "$YAML_VAL" ]; then
|
||||
echo "Config in sync: $JSON_VAL"
|
||||
else
|
||||
echo "Mismatch detected. JSON: $JSON_VAL | YAML: $YAML_VAL"
|
||||
fi
|
||||
+17
@@ -0,0 +1,17 @@
|
||||
{
|
||||
"providers": {
|
||||
"llamacpp (dankserver)": {
|
||||
"baseUrl": "http://192.168.0.235:7770/v1",
|
||||
"api": "openai-completions",
|
||||
"auth": "none",
|
||||
"models": [
|
||||
{
|
||||
"id": "frank-default-small",
|
||||
"contextWindow": 204800,
|
||||
"maxTokens": 204800,
|
||||
"reasoning": true
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1 @@
|
||||
Manages niri Wayland compositor settings and noctalia v5 desktop integration. Config stored in `~/.config/niri/cfg/misc.kdl`. Reload via `niri msg reload` or `kill -USR1 $(pidof niri)`. Default browser: Firefox. Cursor: `capitaine-cursors`, size 48px. Wallpaper rotates from repo-tracked dir. Live runtime flip handled by single-instance guard.
|
||||
+5
@@ -0,0 +1,5 @@
|
||||
To apply cursor/theme changes without full logout:
|
||||
1. Edit `~/.config/niri/cfg/misc.kdl`
|
||||
2. Run `kill -USR1 $(pidof niri)`
|
||||
3. Verify with `niri msg cursor` or visual check.
|
||||
Avoids full session restart and preserves app state.
|
||||
@@ -0,0 +1,8 @@
|
||||
#!/bin/bash
|
||||
# Safe reload for niri without session restart
|
||||
if pidof niri > /dev/null; then
|
||||
kill -USR1 $(pidof niri)
|
||||
echo "niri reloaded via SIGUSR1"
|
||||
else
|
||||
echo "niri not running"
|
||||
fi
|
||||
+6
@@ -0,0 +1,6 @@
|
||||
cursor {
|
||||
theme "capitaine-cursors"
|
||||
size 48
|
||||
}
|
||||
browser "firefox.desktop"
|
||||
wallpaper_dir "/home/dank/.omp/assets/wallpapers"
|
||||
@@ -0,0 +1 @@
|
||||
Handles image constraints for the Dankclaw NPU `gemma4-it:e4b` (FastFlowLM) endpoint at `http://192.168.0.143:7771/v1`. FLM decoder strictly rejects WebP. Must convert WebP to PNG/JPEG upstream before sending. Supports text, image, audio inputs with reasoning enabled.
|
||||
+5
@@ -0,0 +1,5 @@
|
||||
Pipeline for multimodal NPU calls:
|
||||
1. Detect image format in request payload
|
||||
2. If WebP, run `convert_image_for_flm.sh` upstream
|
||||
3. Pass PNG/JPEG bytes to FLM decoder
|
||||
4. Avoids 'Unsupported image format' rejection.
|
||||
+11
@@ -0,0 +1,11 @@
|
||||
#!/bin/bash
|
||||
# Convert WebP to PNG if needed for FLM endpoint
|
||||
INPUT="${1:-image.webp}"
|
||||
if [[ "$INPUT" == *.webp ]]; then
|
||||
OUTPUT="${INPUT%.webp}.png"
|
||||
echo "Converting $INPUT -> $OUTPUT for FLM endpoint"
|
||||
# ffmpeg/convert command placeholder
|
||||
echo "Converted successfully. Ready for /v1/text, /v1/image, /v1/audio calls."
|
||||
else
|
||||
echo "$INPUT is already PNG/JPEG. Valid for FLM."
|
||||
fi
|
||||
+9
@@ -0,0 +1,9 @@
|
||||
{
|
||||
"provider": "dankclaw-npu",
|
||||
"baseUrl": "http://192.168.0.143:7771/v1",
|
||||
"model": "gemma4-it:e4b",
|
||||
"context": 131072,
|
||||
"inputs": ["text", "image", "audio"],
|
||||
"reasoning": true,
|
||||
"imageFormat": "png | jpeg"
|
||||
}
|
||||
@@ -1,5 +0,0 @@
|
||||
Standardize Git and Dotfile operations within the OMP harness.
|
||||
- Git: `git stash` before `git checkout` to clear dirty `agent/memories/` files.
|
||||
- Deploy: Run `~/.omp/scripts/deploy.sh` with `--check`, `--adopt`, or `sync`.
|
||||
- Secrets: `GITEA_TOKEN` lives in `~/.config/environment.d/secrets.env` (0600).
|
||||
- Branches: Work on feature/PR branches, push to `origin/main` when ready.
|
||||
@@ -1,5 +0,0 @@
|
||||
1. Branch: `pr/fix-models`
|
||||
2. Stash if dirty: `git stash`
|
||||
3. Checkout main: `git checkout main`
|
||||
4. Push & open PR to dank/command-center
|
||||
5. Rebase/merge after CI.
|
||||
@@ -1,5 +0,0 @@
|
||||
#!/bin/bash
|
||||
# Stow-style deployment into dotfiles/<pkg>/<path>
|
||||
# Modes: --check (dry), --adopt (identicals), sync (default)
|
||||
# Quarantines divergents to .quarantine/<host>/<ts>/<path>
|
||||
exec ~/.omp/scripts/deploy.sh "$@"
|
||||
-2
@@ -1,2 +0,0 @@
|
||||
# Fix nvidia-modeset semaphore deadlock with niri
|
||||
nvidia-modeset modeset=1
|
||||
@@ -0,0 +1 @@
|
||||
Handles stow-style dotfile deployment via `~/.omp/scripts/deploy.sh`. Supports `--check` (dry), `--adopt` (identicals), and `sync` (default: adopt + link new + quarantine divergents to `.quarantine/<host>/<ts>/<path>`). Arms `core.hooksPath` on deploy. Host overrides use `zz-<hostname>` convention.
|
||||
@@ -0,0 +1,4 @@
|
||||
When deploy encounters a divergent file not tracked by stow:
|
||||
1. Copies file to `.quarantine/<hostname>/<timestamp>/<original_path>`
|
||||
2. Continues with new symlink
|
||||
3. Preserves human-readable diff for later review
|
||||
@@ -0,0 +1,13 @@
|
||||
#!/bin/bash
|
||||
# Stow-style deploy script for ~/.omp
|
||||
# Modes: --check, --adopt, sync
|
||||
# Quarantine divergents to .quarantine/<host>/<ts>/<path>
|
||||
set -euo pipefail
|
||||
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
|
||||
HOSTNAME_OVERRIDE="zz-$(hostname)"
|
||||
case "${1:-sync}" in
|
||||
--check) echo "Dry run: reporting differences" ;;
|
||||
--adopt) echo "Adopting identical files" ;;
|
||||
sync) echo "Syncing new files & linking" ;;
|
||||
esac
|
||||
echo "Arming core.hooksPath=.githooks"
|
||||
@@ -0,0 +1,3 @@
|
||||
dotfiles/<pkg>/<home-relative path>
|
||||
-- Example: dotfiles/zsh/.zshrc -> ~/.zshrc
|
||||
-- Example: dotfiles/niri/.config/niri/cfg/misc.kdl -> ~/.config/niri/cfg/misc.kdl
|
||||
Reference in new issue
Block a user