X Algorithm Optimization
v1.0.0Provides actionable insights into X's For You feed algorithm to optimize content for engagement by predicting and maximizing key user interaction types.
⭐ 0· 71·0 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 (X For You feed optimization) match the delivered assets: comprehensive SKILL.md docs, quick reference, references, and a local evaluator script. There are no unrelated requirements (no cloud creds, no system binaries) and the included evaluate_post.py implements only local heuristic scoring consistent with the skill's goal.
Instruction Scope
SKILL.md and AGENT_USAGE.md instruct agents to read the documentation and synthesize recommendations from the user's context. Instructions remain within content-strategy scope. One caveat: the docs mention optionally refining inferences if you have access to 'internal X data/weights' or personal analytics — that is an optional data source, not required by the skill, but users/agents should not supply internal/privileged platform data unless authorized.
Install Mechanism
No install spec is provided (instruction-only skill). A single Python script is included for local evaluation; it is not downloading or executing remote code, and there are no archive/extract or third-party package downloads present in metadata.
Credentials
The skill declares no required environment variables, credentials, or config paths. The runtime instructions and evaluate_post.py do not read secrets or external env vars. The only external-data suggestion is optional: using your own analytics or internal X data to refine hypotheses — this is a user-supplied data source and not requested by the skill itself.
Persistence & Privilege
The skill is not marked always:true, has no install mechanism that persists code beyond the provided files, and does not instruct changes to other skills or global agent configuration. Autonomous invocation is allowed by platform defaults but is not combined with other red flags here.
Assessment
This skill appears coherent and self-contained: it documents inferred mechanics of X's ranking pipeline and includes a local Python heuristic evaluator. Before installing or running it, consider the following: 1) Inspect scripts/evaluate_post.py yourself — it runs locally and uses only string heuristics, but you should confirm it won't call external network endpoints (it doesn't). 2) Do not feed the skill private or internal X platform data (weights, internal analytics) unless you have explicit authorization and understand legal/ToS implications — the docs mention optional refinement with such data but do not require it. 3) Treat the recommendations as heuristics/inferences, not guaranteed behavior; validate any strategy with your own analytics. 4) Confirm license and provenance if you plan to redistribute (the README claims the source repo is Apache 2.0). If you want extra assurance, request a manual code review focused on any future changes that add network or credential use.Like a lobster shell, security has layers — review code before you run it.
algorithmvk979wekyrbgamcykk0gj4hwwwx83hf8glatestvk979wekyrbgamcykk0gj4hwwwx83hf8gmarketingvk979wekyrbgamcykk0gj4hwwwx83hf8gsocial-mediavk979wekyrbgamcykk0gj4hwwwx83hf8gxvk979wekyrbgamcykk0gj4hwwwx83hf8g
License
MIT-0
Free to use, modify, and redistribute. No attribution required.