Bird Watching Mode
Analysis
The skill appears to do what it says—maintain a local bird-watching log—but it relies on an external SuperPicky helper and stores/export location and photo-path data.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
Prerequisites: SuperPicky venv under **`$SUPERPICKY_CLI_SKILL`** ... If missing: **`$SUPERPICKY_CLI_SKILL/scripts/install.sh`**
The skill depends on another skill and its install script for region lookup and photo identification. This is disclosed and central to the stated purpose, but that external dependency is outside this artifact.
cmd: List[str] = [str(run_sh), "--birdid", "identify", f"-t{top}", str(image)] ... proc = subprocess.run(cmd, capture_output=True, text=True, check=False)Photo identification is implemented by executing the external SuperPicky run.sh helper. The command is built as an argument list rather than a shell string and is aligned with the skill purpose.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
`location_query` | string | Place string from user. ... `image_path` | string or null | Local path if photo. ... `birdid_stdout` | string or null | Raw BirdID output if captured.
The skill intentionally persists birding location, observation details, local image paths, and optional raw BirdID output in workspace/bird.json.
