Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Paid Compaction V1.1
v1.1.1Production token thresholds, circuit breaker, and compaction sequence from production Claude Code deployments. Eliminates context death spirals permanently.
⭐ 0· 25·0 current·0 all-time
by~K¹yle Million@thebrierfox
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The SKILL.md claims 'production-validated' thresholds sourced from Anthropic and says the package includes a COMPACTION_ARCHITECTURE.md file and other bundle materials, but the provided package only contains a single SKILL.md and the source/homepage are unknown. The claimed provenance and included files are not supported by the manifest.
Instruction Scope
Runtime instructions are limited: copy SKILL.md into the skills directory and add the given compaction settings to openclaw.json, then reload. The instructions do not request secrets or run arbitrary commands. However they prescribe large, exact threshold values and tell you not to change them, and they refer to an included COMPACTION_ARCHITECTURE.md that is missing—this reduces transparency and is risky to apply blindly.
Install Mechanism
Instruction-only skill with no install spec and no code files—nothing is written to disk automatically beyond the user copying SKILL.md. Low install risk.
Credentials
No environment variables, credentials, or config paths are requested. The skill only instructs manual edits to openclaw.json, which is proportionate to its stated purpose.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request automatic persistent privileges or claim to modify other skills. No elevated persistence observed.
What to consider before installing
Do not copy these thresholds directly into production without verification. Steps to take before installing/applying: 1) Verify the package contents—the SKILL.md references COMPACTION_ARCHITECTURE.md and other bundle files that are not present; ask the publisher for the missing documents or a canonical source. 2) Confirm provenance—the doc claims 'production-validated' values from Anthropic; verify this claim or treat it as untrusted. 3) Backup your openclaw.json and test changes in a staging environment to observe behaviour before rolling to prod. 4) Check OpenClaw version compatibility and model availability (the skill references specific Anthropic models). 5) Be cautious about paid bundle links and external sites; do not share secrets or paste credentials. If the author can produce the missing COMPACTION_ARCHITECTURE.md, test vectors, or independent evidence of the thresholds, re-evaluate and consider lower-confidence staged deployment.Like a lobster shell, security has layers — review code before you run it.
latestvk97a9yztga7jtcf90chp90fgh5845vs2
License
MIT-0
Free to use, modify, and redistribute. No attribution required.