Context Anchor
PassAudited by ClawScan on May 1, 2026.
Overview
This skill coherently runs a local script to read scoped memory/context markdown files and summarize them, with no evidence of network access, credential use, or file modification.
This appears safe for its stated purpose if you are comfortable with the agent reading your workspace memory and context markdown files. Be cautious about adding it to every-session startup unless those files are curated and trusted.
Findings (2)
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.
Private, outdated, or untrusted notes in those files could influence what the agent does next or be surfaced in the briefing.
The skill is designed to read persistent workspace memory and active context files and place selected contents back into the agent's current context.
`memory/current-task.md`, `memory/YYYY-MM-DD.md`, `context/active/*.md`
Use this with workspace memory files you trust, and review the referenced memory/context files before relying on them for important decisions.
Installing and using the skill means allowing this local script to run, though the provided code only shows scoped reads and terminal output.
The skill operates by running an included bash script that reads local markdown files using standard shell tools; this is disclosed and central to the skill's purpose.
#!/bin/bash ... cat "$task_file" ... grep -n -i -E ... "$file"
Run it only from the intended workspace and keep the script source reviewable if you modify or redistribute it.