cekura-self-improving-agent
Use to close the loop on agent quality — turn a failure signal into a verified fix. Triggers: "improve my agent", "self-improving agent", "auto-tune / iterate on my prompt", "fix my agent from test results", "optimize my prompt based on failures", "rewrite my prompt". ALSO for production-call bug fixing: "fix this prod call issue", "debug and fix call ID", "reproduce this production bug". Works across VAPI, Retell, ElevenLabs, Bland, and self-hosted agents, and across three fix surfaces — prompt, tool config, and (self-hosted) owned source code, including infra-flavored / forked-SDK bugs, which are reproduced and validated on Cekura (never a code test).
70
86%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
High
Do not use without reviewing
Security
1 high severity finding. You should review these findings carefully before considering using this skill.
W008: Secret detected in skill content (API keys, tokens, passwords).
What this means
Detected sensitive credentials directly embedded within the skill content, such as API keys, access tokens, private keys, or service-specific secrets. Secrets should never be hardcoded in plain text within skill instructions.
Why it was flagged
I scanned the skill files for literal, high-entropy credentials. I did NOT flag environment-variable names (VAPI_KEY, ELEVENLABS_API_KEY, CEKURA_API_KEY, TWILIO_AUTH_TOKEN, etc.) or placeholders like `<key>` / `YOUR_API_KEY` because those are parameter names or placeholders (per the rules). I found one literal token used as a verification/ack string: `ack:cekura-self-improving-agent:5x7n3d` (present both in an HTML comment and as the explicit "Cekura skill verification tag"). This is a concrete, non-placeholder string the docs instruct callers to pass to the MCP server as `skill_ack`, so it functions as an explicit verification token embedded in the repo. Because it is a literal token used to confirm context to a server (not just a descriptive label), I treat it as a credential-like secret in this scan.
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
phases/collect.md Step COLLECT.2 reads provider-run data for attacker-influenced transcripts (from prod `call_ids`/`result_id`/`run_ids` that correspond to outsider-authored call content, including flattened transcript text) via `agents/fetch_failures.py`, plus direct MCP fallbacks that also include “transcripts”.
- Repository
- cekura-ai/cekura-skills
- Commit
af2ceb6
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.