Workspace Context Linter

Diagnose always-loaded workspace context files such as AGENTS.md, SOUL.md, USER.md, MEMORY.md, and TOOLS.md. Use when you want to reduce context bloat, detec...

Security Scan

The skill is internally consistent: it inspects local workspace markdown files to produce a diagnostic report, it does not request credentials or network access, but the included linter script as provided appears to be truncated/buggy and should be reviewed before running.

Purpose & Capability

The name/description match the observed behavior: the code and SKILL.md both load and analyze core workspace files (AGENTS.md, SOUL.md, USER.md, MEMORY.md, TOOLS.md) to detect duplicates, overweight sections, and misplaced content. The required resources (none) are proportionate to that purpose.

Instruction Scope

SKILL.md instructs the agent to read the included reference files and to run scripts/context_linter.py with a local scope. The script reads files from Path.cwd() and optionally memory/*.md or custom paths, which is coherent with the stated purpose but means the tool will read whatever files are present in the agent's working directory. That behavior is expected for a linter but you should be aware it will examine local workspace files (not remote endpoints).

Install Mechanism

There is no install spec (the skill is instruction + a Python script). No packages are downloaded or extracted and there are no external URLs invoked in the code. Running the skill requires executing the provided Python script locally.

Credentials

The skill requests no environment variables, no credentials, and no config paths. The script only reads local workspace files and has no network or secret-access code, so requested access is proportional to the purpose.

Persistence & Privilege

The skill is not always-loaded (always: false) and is user-invocable. It does not attempt to persist settings, modify other skills, or require elevated privileges. Autonomous invocation is allowed by default on the platform but that is not combined with other concerning attributes here.

Assessment

This skill appears to do what it says: a local Python linter that diagnoses always-loaded workspace markdown files. It does not request credentials or talk to external servers. Two precautionary points before you run it: (1) the provided script snippet appears to be truncated and contains at least one likely bug (an incomplete render/suggestions section and a probable undefined variable), so running it as-is may produce a crash — inspect the full scripts/context_linter.py file and fix/complete it before executing; (2) the linter reads files under the current working directory (and optionally memory/*.md or custom paths), so run it in a repository or folder that contains only files you are comfortable having inspected. If you want extra safety, open the full script and verify there are no network calls or exec/system calls (the visible code uses only standard library file reads and text analysis).

Audit core workspace context files without rewriting them.

Core workflow

1. Load the core context files that exist.
2. Summarize each file's likely role.
3. Detect duplicate rule themes, overweight sections, and misplaced content.
4. Produce a text report with priorities and suggested moves.

Read references as needed

• Read references/rules.md for the lint categories and severity model.
• Read references/report-format.md for the output structure.
• Read references/file-roles.md for what each core context file should usually contain.
• Read references/move-guidelines.md when deciding where content should move.
• Read references/release-minimal.md before packaging or publishing so the first public surface stays minimal.

Use scripts as needed

• Use scripts/context_linter.py [--scope core|core+memory|custom] [--paths ...] [--output report.txt] to run the linter.

Operating rules

• Prefer diagnosis over auto-editing.
• Treat duplicates as a maintenance problem unless they create real execution ambiguity.
• Treat overweight sections as candidates for extraction, not automatic deletion.
• Keep the first version focused on core context files and text reports.