RedHat File Operations

Perform safe, auditable file operations with path validation and backups.

Security Scan

The skill's declared purpose (safe, auditable file operations) matches its instructions and requirements; it is an instruction-only skill that does not request credentials or install code, but it performs destructive file actions so operator controls are still important.

Purpose & Capability

Name, description, and runtime instructions are aligned: the skill describes reading/writing/updating/deleting/moving/copying files with path validation and backups and does not request unrelated binaries, env vars, or external services.

Instruction Scope

SKILL.md stays within the stated purpose and contains explicit safety rules (preflight checks, resolved path printing, backups). It does not instruct network transmission or credential access. One ambiguity: it refers to an 'MCP-enabled environment' and 'workspace root(s)' but does not declare how those roots are obtained or enforced — that platform integration detail is external to the skill and should be verified by the deployer.

Install Mechanism

Instruction-only skill with no install spec and no code files. Nothing is written to disk by the skill itself during installation.

Credentials

The skill declares no environment variables, credentials, or config-path requirements. It does not request broad secrets or unrelated access.

Persistence & Privilege

always:false (normal) and user-invocable:true. The skill can be invoked autonomously by the agent (disable-model-invocation:false) — this is platform default, but combined with file-editing capabilities it means you should confirm agent policies and prompting/confirmation behavior before allowing autonomous runs.

Assessment

This skill appears coherent with its stated goal and does not request credentials or install code, but it performs potentially destructive file operations. Before installing or enabling it: (1) confirm what the platform supplies as the 'workspace root' and ensure the skill cannot escape that scope, (2) require explicit user confirmation for delete/recursive/overwrite actions and test on non-production data, (3) ensure backups created by the skill are stored where you expect and protected, (4) if you are uncomfortable with autonomous invocation, disable autonomous invocation or require manual invocation/approval in your agent policy, and (5) review audit/logging policies so all file changes are recorded. If you need more assurance, ask the maintainers to specify how workspace roots are determined and to provide an implementation review or sample run logs.

SKILL: file-operations

Purpose

Perform safe, auditable file operations (read/write/update/delete/move/copy) in an MCP-enabled environment with path validation and backups.

When to Use

• Any task requires editing files on disk.
• You need to ensure changes are safe and reversible.
• You must avoid accidental overwrites or destructive operations.

Inputs

• operation (required, enum: read|write|update|delete|move|copy).
• paths (required, object):
  • source (optional, string)
  • target (optional, string)
  • targets (optional, string[])
• content (optional, string): for write/update.
• backup (optional, boolean, default: true for overwrite/delete).
• constraints (optional, string[]): e.g., “only within workspace”.

Steps

1. Resolve absolute paths and normalize.
2. Validate scope:
   • operation stays inside the intended workspace root(s)
   • block path traversal/suspicious targets
3. Preflight:
   • confirm existence for read/update/delete
   • confirm non-existence (or allow overwrite explicitly) for write/move
4. If overwrite/delete and backup=true, create a backup copy (timestamped).
5. Execute the operation with explicit reporting (what changed).
6. Post-validate:
   • file exists where expected
   • content/diff matches expectation

Validation

• Operation is reversible when possible (backup exists for destructive actions).
• The final target path is verified before execution.
• The operation result is explicitly reported (not silent).

Output

operation: "<operation>"
affected: ["<path>"]
backup_paths: ["<path>"]
result: "success|blocked|failed"
notes: ["..."]

Safety Rules

• Never delete or overwrite recursively without explicit user confirmation.
• Never operate on computed paths without printing the resolved final paths first.
• Prefer patch-based updates over full-file rewrites when possible.

Example

Update a config file:

• operation: update
• paths: { target: "apps/api/.env.example" }
• Validation: backup created; diff is minimal; file still parses.