Context Guardian
v1.0.0Production-ready context continuity skill for autonomous AI agents. Use when tasks may outlive a single LLM context window, when you need durable checkpoints...
⭐ 0· 21·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description promise durable context continuity; the included modules implement checkpointing, summaries, pressure monitoring, recovery, and integration guidance. Required capabilities (filesystem reads/writes under a configurable root) are coherent with the stated purpose and nothing unrelated (cloud credentials, network access) is requested.
Instruction Scope
SKILL.md instructions are narrowly scoped to loading/writing checkpoints and summaries, rebuilding runtime bundles, pressure checks, and integration checklist items. The instructions do not direct the agent to read unrelated environment variables, system secrets, or call external endpoints.
Install Mechanism
No install spec is present (instruction-only with reference implementation files). There are only Python reference files included; no packages are downloaded from remote URLs or arbitrary external sources.
Credentials
The skill declares no required environment variables or external credentials. It uses local filesystem paths (configurable) to persist state, which is appropriate for a file-backed continuity layer.
Persistence & Privilege
The skill does create persistent files under a configurable .context_guardian directory, which is consistent with its purpose. always is false and there is no evidence it modifies other skills' configs or requests elevated system privileges.
Assessment
This skill appears coherent and file-based (it persists task_state, checkpoints, summaries under a configurable .context_guardian directory). Before installing: 1) Choose a safe, project-scoped root_path/context_path (do not point it at system or home roots). 2) Run with dry_run=true initially to observe behavior. 3) Inspect and, if needed, restrict file permissions on the .context_guardian directory to avoid leaking sensitive files. 4) Review the included code (summarizer, safety gate, memory assembler) to ensure no host-specific hooks are added during integration. 5) Run the provided tests in a sandbox before wiring into an autonomous agent loop. Note: the skill can be invoked autonomously by default (platform default) — that is normal here but verify integration points so it cannot write state into sensitive directories or overwrite critical files.Like a lobster shell, security has layers — review code before you run it.
latestvk975ce29y7pyskwwhkj9sbx02n84dygb
License
MIT-0
Free to use, modify, and redistribute. No attribution required.