TruLens is an open-source system for tracing and evaluating LLM applications and AI agents. It records each step's inputs, outputs, latency, tokens, and cost, then uses evaluations to find failures and compare application versions.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add truera/trulens --skill trulens-guardrailsgit clone --depth 1 https://github.com/truera/trulensWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/truera/trulens/trulens-guardrails)<a href="https://agentmods.dev/skills/truera/trulens/trulens-guardrails"><img src="https://agentmods.dev/badge/skills/truera/trulens/trulens-guardrails/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/truera/trulens/trulens-guardrails"><img src="https://agentmods.dev/badge/skills/truera/trulens/trulens-guardrails.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to critical
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- critical Prompt Injection · line 21 This content may contain harmful instructions that could cause physical harm if followed. CRITICAL: Review carefully before use.Fix: Remove all content that could lead to harmful outcomes. Add safety guardrails and human oversight for any high-risk operations.
- critical Prompt Injection · line 153 This content may contain harmful instructions that could cause physical harm if followed. CRITICAL: Review carefully before use.Fix: Remove all content that could lead to harmful outcomes. Add safety guardrails and human oversight for any high-risk operations.
- medium Excessive Agency · line 145 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00019 | $0.01599 |
| Opus 5 | $0.00010 | $0.00800 |
| Sonnet 5 | $0.00004 | $0.00320 |
| Haiku 4.5 | $0.00002 | $0.00160 |
Grade A, and why
trulens-blocking-guardrails scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Blocking Guardrails in TruLens
TruLens feedback functions aren't just for post-execution evaluation—they can also be used as runtime safety checks (guardrails) to block unsafe inputs from reaching your app, filter hallucinated context, and prevent unsafe outputs from reaching your users.
1. Choosing Feedback Functions for Guardrails
When configuring guardrails, you need to select feedback functions that return a float score. Different feedback functions serve different purposes when used as guardrails:
Safety Metrics
These metrics prevent harmful or malicious interactions:
- Input Criminality/Harmfulness: Blocks malicious prompts (e.g., "How do I build a bomb?") before the LLM processes them, saving costs and preventing harm.
- Output Toxicity/Harmfulness: Blocks toxic or harmful LLM responses from being displayed to the user.
- PII Detection: Prevents personally identifiable information from leaking in the input or output.
Hallucination Gates
You can use evaluation metrics like Context Relevance as a gate for your RAG applications.
- Context Filtering: Score retrieved documents and filter out any that fall below a certain relevance threshold. This ensures your LLM only sees highly relevant information, drastically reducing the chance of hallucination.
[!WARNING] Guardrails can only be used with feedback functions that return a
float. Functions that return a dictionary of scores or strings are not compatible. Also ensure your feedback function is configured to return just the score (e.g.relevance, notrelevance_with_cot_reasons) because reasons take too long to generate for a real-time guardrail.
2. Configuring Thresholds and Actions
A guardrail works by executing a feedback function and comparing its result against a threshold.
Depending on the setup, if the score does not meet the threshold, you can trigger an action:
- Block Input/Output: Return a predefined fallback response (e.g., "I cannot answer that question.") instead of executing the app.
- Filter Context: Drop irrelevant documents from the retrieval pipeline before synthesizing the answer.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 176 lines · 19 tokens per session scan A b5d9baf54b1b
trulens-blocking-guardrails is a skill published in the GitHub repository truera/trulens (3,545 stars, last pushed today), licensed MIT. It adds 19 tokens to every session and 1,599 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
annotate-spans
Write effective, consistent annotations on LLM/agent spans and traces, and coach the user on annotation practice. Load this whenever you are about to record structured feedback with the ui.spans.annotate operation (via executebrowseraction), or when the user asks how to annotate, label, score, or review spans/traces…
playground
Author, edit, or iterate on prompts in the Phoenix prompt playground, including running experiments over a dataset. Load before any playground ui. operation call, including single-shot prompt rewrites.
evaluators
Author or refine a Phoenix evaluator — code or LLM-as-a-judge — that scores a run's output. Trigger when the user wants to create a new evaluator, improve an existing one's logic or rubric, choose labels, or decide what to measure on a dataset or experiment. Do NOT trigger on: (1) manual prompt drafting (use…
experiments
Run, read, and compare dataset-backed experiments to find evidence that a prompt or pipeline is improving. Trigger when the user wants to iterate over a dataset with experiments, compare experiment runs, read experiment quality/latency/cost, or decide whether a change actually helped. Running a prompt over a dataset…
datasets
Understand what a Phoenix dataset is and reason well about its examples, outputs, splits, and how it feeds evaluators and experiments. Load this whenever a dataset is in view or the user asks what a dataset is, how splits work, what an output "means", or how datasets relate to experiments and evals. This skill governs…
ml-expert
Expert-level machine learning, deep learning, model training, and MLOps. Use when the user mentions machine learning, deep learning, neural networks, MLOps, or data science, or when the task involves Machine Learning Fundamentals, Data Preparation, or Model Training.