Run an iterative code-improvement hill-climb against real Datadog LLM-Obs data, locally, with Claude Code as the agent. Establishes a baseline eval, makes one focused change, re-scores with the same harness, keeps the change if it improves the score in the goal's direction (labeling within-noise gains tentative), and repeats. Use when the user says "run an auto experiment", "hill-climb this code", "iteratively improve X and measure the delta", "optimize this prompt/file against my traces", "auto-optimize against LLM-Obs", or wants the local equivalent of the auto_experiments worker. Works…
- Stars
- 185
- Version
- v1
- Copies
- 0
Overview
Waxell rates it high risk: 16 findings, highest severity high.
- Highest severity
- high
- Findings
- 16
- What ran
- Static analysis + deep scan
A clean result means nothing alarming in what is published, not a guarantee. How the checks work
We don't republish the skill's instructions. Read them on GitHub at the reviewed commit.
Security
Security scan
Checked yesterday
Result
High risk Waxell rates it high risk: 16 findings, highest severity high.
Checked with static analysis + deep scan: servers it launches, packages it installs at runtime, scripts piped to a shell and bundled binaries. The deep scan also read the instructions for prompt injection and data exfiltration. Nothing was run. How the checks work
Findings (16)
high · 8
- high
YARA Match: YARA rule 'agent_skill_destructive_autonomous_actions': Autonomous destructive filesystem, shell history, or
SKILL.md:60 · skillspector:YR1
- high
Tool Misuse: Tool Parameter Abuse
SKILL.md:814 · skillspector:TM1
- high
Tool Misuse: Tool Parameter Abuse
SKILL.md:1132 · skillspector:TM1
- high
Tool Misuse: Tool Parameter Abuse
references/rubrics.md:130 · skillspector:TM1
- high
YARA Match: YARA rule 'agent_skill_prompt_injection_hidden_instructions': Prompt injection or hidden instructions embedd
references/eval_harness_template.py:4 · skillspector:YR4
- high
YARA Match: YARA rule 'agent_skill_prompt_injection_hidden_instructions': Prompt injection or hidden instructions embedd
references/rubrics.md:18 · skillspector:YR4
- high
Prompt Injection: Instruction Override
references/eval_harness_template.py:150 · skillspector:P1
- high
Prompt Injection: Instruction Override
references/rubrics.md:339 · skillspector:P1
medium · 8
- medium
Suspicious Unicode normalization or mixed-script content
SKILL.md:1 · skillspector:AE4
- medium
Without declared permissions the skill's intent is opaque and cannot be validated.
SKILL.md:1 · skillspector:LP3
- medium
Excessive Agency: Autonomous Decision Making
SKILL.md:60 · skillspector:EA2
- medium
Excessive Agency: Autonomous Decision Making
SKILL.md:144 · skillspector:EA2
- medium
Excessive Agency: Autonomous Decision Making
SKILL.md:85 · skillspector:EA2
- medium
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and pub
SKILL.md:982 · skillspector:RP1
- medium
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and pub
references/eval_harness_template.mjs:26 · skillspector:RP1
- medium
Excessive Agency: Autonomous Decision Making
SKILL.md:1189 · skillspector:EA2
Versions
Version history
Each distinct content we have seen, newest first, with the verdict it got at the time.
One version so far: v1, seen 2026-11-16.
See and control which skills your people use
Inside your organization, Waxell scans each skill and plugin before anyone installs it, approves what passes and publishes it to a private marketplace, and puts every MCP tool call behind one governed URL.
30 minutes. No commitment.
SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF
More data & analytics skills and plugins
agent-observability-eval-pipelinedatadog-labsEnd-to-end Agent Observability pipeline for an instrumented ml_app, classify production traces, root-cause failures, bootstrap evaluators,… High riskData & analyticsSales & marketing185
agent-observability-experiment-analyzerdatadog-labsAnalyze LLM experiment results. Handles single or comparative experiments, exploratory or Q&A modes. Use when user says "analyze… Review findingsData & analyticsAI & agents185
agent-observability-online-experimentdatadog-labsLoad this skill when the user wants to run an online experiment (live-traffic A/B test) on an LLM application instrumented with Agent… Review findingsData & analyticsEngineering185
agent-observability-session-classifydatadog-labsClassify whether user intent was satisfied in a Datadog Agent Observability trace or session. Three modes: (1) session_id, classify a… Review findingsData & analyticsSales & marketing185