OpenClaw skill for using galdr's ARC workflow to turn local audio files into grounded, time-ordered listening-experience prompts backed by timed audio-derived proxies for pattern, attention, heard pressure, surface balance/evidence, harmony, melody, overtones, and silence/re-entry structure, plus report-level pulse evidence. These are interpretation cues, not measured listener responses. Use when asked to analyze a song, explain what makes a track work structurally, generate a listening experience.
galdr
Use this skill when an OpenClaw agent needs to analyze music from a local audio file and produce a grounded listening-experience prompt from measurable audio structure.
galdr is a music perception CLI for AI agents. Its default workflow is ARC: analyze a track into time-ordered listener-state traces, then assemble those traces into a prompt for grounded listening-experience prose. The metrics are evidence. The ARC prompt is the main user-facing output.
Important: skill vs CLI
Current OpenClaw CLI install command:
openclaw skills install galdr
ClawHub may display an owner-qualified command such as openclaw skills install @sellemain/galdr. As of OpenClaw 2026.6.8, the released CLI expects the bare skill slug galdr.
Installing this skill teaches OpenClaw how to use galdr. It does not install the galdr command itself.
This skill packaging revision is 0.8.0 and targets the separately installed galdr CLI 0.8.0. The skill is licensed MIT-0; the CLI remains MIT-licensed.
The PyPI wheel contains the runtime CLI/library and bundled prompt templates. The OpenClaw skill is distributed separately through ClawHub so agent instructions can stay a clean skill artifact instead of being installed as Python package data.
Before starting:
galdr --version
If missing, install the CLI from a trusted source:
pip install galdr
# or from source:
git clone https://github.com/sellemain/galdr.git
cd galdr
pip install -e .
Preferred trusted sources:
If provenance matters, verify the PyPI metadata or install from the source repository before running it.
When to use this skill
Use galdr when the user asks to:
- analyze a local audio recording
- describe what makes a track work structurally
- generate a grounded listening experience
- create an evidence packet for another model to write from
Do not use galdr for:
- general music trivia
- ordinary recommendation lists
- pretending the metrics prove private emotional intent
- acquiring source media; the caller supplies a local audio file
OpenClaw agent contract
Prefer the ARC path unless the user explicitly asks for raw metrics, debugging, or agent-internal traces.
Default sequence:
- Confirm the supplied local audio file is readable.
- Run
galdr listento generate listener-state traces. - Assemble the ARC prompt with
--template arc. - Review the prompt.
- Write the listening experience yourself or pass the prompt to the requested model.
The stream is evidence. Walk the track through time before summarizing. Do not invent emotional claims that the structure does not support.
When a user wants to configure another agent or project to use Galdr, point them to docs/AGENT-ONBOARDING.md and examples/agent/AGENTS.md. Those files contain the public, copy-pasteable agent instruction path.
Use the ARC prompt family when the user asks for a specific reading mode:
galdr assemble my-track --template arc-family --lens sound > sound.txt
galdr assemble my-track --template arc-family --lens dance > dance.txt
galdr assemble my-track --template arc-family --lens meaning > meaning.txt
galdr assemble my-track --template arc-family --lens structure > structure.txt
galdr assemble my-track --template arc-family --lens classical > classical.txt
galdr assemble my-track --template arc-family --lens ritual > ritual.txt
Lens guide:
default— general public listening pagesound— sound as physical shape, pressure, density, space, body, and motiondance— movement contract: groove, repetition, build/drop, and bodily usestructure— compact mechanical/form witnessmeaning— human situation carried by soundclassical— instrumental/classical/large-form attention over timeritual— private ritual reading with weak-fit boundary behavior
Core Workflows
Local audio file → ARC prompt
galdr listen "audio/track.flac" --name my-track
galdr assemble my-track --template arc > prompt.txt
Galdr does not download audio, videos, lyrics, or captions. galdr doctor checks the local Python and ffmpeg/ffprobe setup. Existing analysis and manually supplied background can still be read; legacy lyric and visual fields are ignored.
Raw second-by-second analysis (advanced)
Galdr is strongest when read as a time-ordered listener-state trace. The stream is the primary evidence. Whole-track interpretation comes after walking the track through time.
Start with:
analysis/<slug>/<slug>_stream.jsonanalysis/<slug>/<slug>_perception.jsondocs/PERCEPTION-MODEL.md
Useful extras:
*_harmony_stream.json*_melody_stream.json*_overtone_stream.json*_report.jsongaldr assemble <slug>
Reading order:
- Read
PERCEPTION-MODEL.mdfirst. - Treat
*_stream.jsonas the main evidence surface. - Walk the track in order.
- Mark transitions: silence, re-entry, pattern breaks, attention shifts, pressure-state changes, harmonic movement.
- Translate pressure fields into listening language: comes forward, holds, releases, empties. Do not quote LUFS values in experience prose.
- Only then compress upward into a larger interpretation.
Do not:
- jump straight to a whole-song mood summary
- treat summary metrics as more important than the stream
- ignore silence/re-entry structure
- overclaim emotional certainty from structure alone
- quote loudness/LUFS readings as if they were the experience
Minimal recipe:
galdr listen track.wav --name my-track
jq '.stream[0:12]' analysis/my-track/my-track_stream.json
jq '.summary' analysis/my-track/my-track_perception.json
galdr assemble my-track > prompt.txt
Send the ARC prompt to another model
Only do this if the operator explicitly wants model-written prose. Review the assembled ARC prompt before piping it to claude, llm, or any other external model endpoint.
galdr assemble my-track --template arc | claude
galdr assemble my-track --template arc | llm
Optional Python agent pattern
from galdr import listen
analysis = listen("audio/track.flac", name="my-track")
prompt = analysis.to_prompt(template="arc")
# Review prompt before sending it to an external model.
Context and template inputs
Assembly includes audio-derived evidence by default. For repeatable supplied metadata,
background, or custom data, use --context-file notes.json; the JSON object is labeled
as caller-supplied, not Galdr measurement. For one-off work, combine context with the
assembled output in your agent. Use --template path/to/prompt.md for custom prompt
instructions. Do not quietly promote supplied claims to audio evidence.
--template arc prepends the default listening-experience rules: tone, format, interpretation bounds, and the instruction to walk the track through time. --template arc-family --lens <name> uses the shared prompt-family base plus one deliberate reading lens. Omit templates only when you want a raw data block.
Interpreting galdr output
ARC is the default output path. The metrics exist to keep that prose grounded: use them as evidence for what changes, returns, releases, locks, or breaks over time.
See references/metrics.md for full metric reference.
Quick read:
patternnear 1.0 → little detected disruption; lower values → more beat/spectral/energy disruption, not proof of a listener’s statesurface_balancenegative → harmonic-side energy dominant; positive → percussive-side energy dominant. HPSS balance does not identify voice or instruments; confirm heard sources by listening.- This signed rule applies to stream/ARC and
surface_balance_signed; legacy report/catalogsurface_balanceis an unsigned percussive fraction. chroma_motion_semantics: absolute_cosine_distance_v1marks absolute adjacent-frame motion; untagged older values were track-relative and must not be compared directly.pitch_gridis 12-class chroma concentration, not tuning accuracy;overtone_fitcombines matched-slot coverage and closeness, not purity.harmonic_pullis normalized to each track’s own peak; read local movement, not absolute dissonance or cross-track tension.- Pressure summary shares count active
pressure_statelabels, excluding silence states/flags and unknown; ARC spans use a coarser classifier. - Top-level stream
pressure_stateand active-frame summary percentages → pressure-motion labels; nestedsurface_evidence.pressure_statedescribes thin/held/pressurized surface texture instead pattern_breaksis a typed event list (time/type); clustered candidate breaks or silences near the end may support a release reading after audio confirmationsilencescarrystart/end/duration/relativedepth_db;silence_reentriesseparately carrypost_attention/recovered. These are algorithmic candidates, not proof of listener re-lock
Writing ARC experience prose yourself
When writing experience prose yourself from galdr evidence, prefer galdr assemble <slug> --template arc. If you are writing from raw assembled output without the template:
- First-person listener perspective, present tense
- Timestamps only at structural pivots: silences, pattern breaks, major energy shifts
- Translate metrics; describe what they mean, do not quote numbers
- LUFS/pressure values are evidence, not prose; write “pressure comes forward / holds / releases / empties”
- Body anchors such as chest, jaw, sternum sparingly; two or three for the whole piece
- End at the final sound event; no aftermath, no reflection
- Around 800 words, no section headers
Other commands
galdr doctor # inspect local audio decoder health
galdr catalog # local analysis index (operator tooling)
galdr catalog --track NAME # summary card for one track
Related skills
- Downloads
- 2.7kAll time
- Security audit
- Pass
- Last updated
- 1mo ago
- Current version
- v0.8.0
- License
- MIT-0
