agentleFS
Seeed-ProjectsSeeed-Projects/Seeed-Jetson-DevelopTool/skills/codex/cochl-sense-setup/AGENTS.md

Seeed-Jetson-DevelopTool / cochl-sense-setup

AGENTS.md57 starsChanged 4 months agoMITView on GitHub

Deploy Cochl.Sense sound recognition on Jetson (37 sounds, 94% F1 score). Requires a trial license from Cochl and a USB microphone. Supports JetPack 4.6+.

  • Reads credentials
  • Installs packages

The file

We split this AGENTS.md into the rules it gives an agent

7 rules. About 1 is common, 4 are distinctive, and 2 are very specific to this project.

  • ≈1

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run one phase at a time. After each phase:

    line 20
  2. Very specific:

    If output contains [STOP] → stop immediately, consult the failure decision tree

    line 21
  3. Very specific:

    If output ends with [OK] → tell the user "Phase N complete" and proceed to the next phase

    line 22
  4. Distinctive:

    After approval, create a project at: https://dashboard.cochl.ai

    line 32
  5. Common:

    If yes: ask the user to provide the API key now — it is needed in Phase 3.

    line 35
  6. Distinctive:

    Verify both are present:

    line 58
  7. Distinctive:

    Check USB microphone is detected:

    line 94

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Cochl.Sense Sound Recognition — Jetson Setup

Cochl.Sense is a machine listening SDK that recognises 37 target sounds at 94% F1 score. It runs on Jetson via a pre-built AArch64 C++ SDK. A trial license is required before any code can execute — ask the user if they have one before starting Phase 2.

Hardware: Jetson with JetPack 4.6+, USB microphone (16-bit, 22050 Hz)

Reference: references/source.body.md (Cochl docs, SDK download links, tutorial repo)


Execution model

Run one phase at a time. After each phase:

  • If output contains [STOP] → stop immediately, consult the failure decision tree
  • If output ends with [OK] → tell the user "Phase N complete" and proceed to the next phase

Phase 1 — license & project (~5 min, human action required)

Ask the user: "Do you already have a Cochl.Sense trial license and a project API key?"

If no:

  1. Request a trial license: https://forms.gle/Pa2iYWirLJVNS7Pp6
  2. After approval, create a project at: https://dashboard.cochl.ai
  3. Copy the API key from the project dashboard

If yes: ask the user to provide the API key now — it is needed in Phase 3.

[OK] when the user has confirmed they have an API key.


Phase 2 — download SDK & tutorials (~5 min)

Check JetPack version:

cat /etc/nv_tegra_release

Confirm JetPack 4.6 or later (L4T R32.7+). If earlier → [STOP] unsupported JetPack.

Download the AArch64 SDK from Cochl Docs → Resources section. (Exact download URL is in references/source.body.md.)

Clone the tutorials repository:

git clone https://github.com/cochlearai/sense-sdk-cpp-tutorials

Verify both are present:

ls sense-sdk-cpp-tutorials-main/
ls sense/   # the downloaded SDK folder

[OK] if both directories exist.


Phase 3 — setup: move SDK into tutorials (~1 min)

Move the sense/ SDK folder into the tutorials directory:

mv sense/ sense-sdk-cpp-tutorials-main/

Verify the layout:

ls sense-sdk-cpp-tutorials-main/sense/

Expected: include/, lib/, and related SDK files.

Insert the API key into the sample config (path may vary — check references/source.body.md):

# Edit the config file and replace YOUR_API_KEY with the actual key
sed -i 's/YOUR_API_KEY/<API_KEY>/' sense-sdk-cpp-tutorials-main/config.json

[OK] when SDK is in place and API key is set.


Phase 4 — build & run (~3–5 min)

Check USB microphone is detected:

arecord -l

Expected: at least one capture device listed. If none → [STOP] no microphone detected.

Build the sample application:

cd sense-sdk-cpp-tutorials-main
mkdir -p build && cd build
cmake ..
make -j$(nproc)

Run the sample:

./sense_sample

Expected output: a stream of recognised sound labels printed to stdout (e.g. Speech, Silence).

[OK] if sound labels appear in the output.


Failure decision tree

SymptomAction
[STOP] unsupported JetPackCochl.Sense requires JetPack 4.6+. Flash a supported version before continuing
User does not have a licenseDirect them to https://forms.gle/Pa2iYWirLJVNS7Pp6 — cannot proceed without it
SDK download failsCheck Cochl Docs → Resources for updated link; see references/source.body.md
git clone failsCheck network connectivity: ping github.com. Retry or download ZIP manually
cmake not foundsudo apt install cmake
Build fails — missing headersConfirm sense/ is inside sense-sdk-cpp-tutorials-main/ and re-run cmake
[STOP] no microphone detectedRun lsusb to confirm USB mic is connected. Try a different USB port. Check `dmesgtail -20`
sense_sample exits with auth errorAPI key is wrong or expired. Re-check the key in the dashboard at https://dashboard.cochl.ai
Output shows only SilenceMicrophone gain too low. Run alsamixer and increase capture volume

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.